Control systems and methods for HVAC equipment with fluid loop thermal energy storage

Abstract

Disclosed herein are related to a system, a method, and a non-transitory computer readable medium for operating an energy plant. In one aspect, the system generates a regression model of a produced thermal energy load produced by a supply device of the plurality of devices. The system predicts the produced thermal energy load produced by the supply device for a first time period based on the regression model. The system determines a heat capacity of gas or liquid in the loop based on the predicted produced thermal energy load. The system generates a model of mass storage based on the heat capacity. The system predicts an induced thermal energy load during a second time period at a consuming device of the plurality of devices based on the model of the mass storage. The system operates the energy plant according to the predicted induced thermal energy load.

Description

CROSS-REFERENCE TO RELATED PATENT APPLICATION

where θ*_HL contains the optimal high level decisions (e.g., the optimal load {dot over (Q)} for each of subplants) for the entire prediction period and J_HL is the high level cost function.

To find the optimal high level decisions θ*_HL, the asset allocator 445 may minimize the high level cost function J_HL. The high level cost function J_HL may be the sum of the economic costs of each utility consumed by each of subplants for the duration of the prediction time period. For example, the high level cost function J_HL may be described using the following equation:
J_HL(θ_HL)=Σ_k=1^nhΣ_i=1^ns[Σ_j=1^nut_s·c_jku_jik(θ_HL)]  Eq. (2)
where n_h is the number of time steps k in the prediction time period, n_s is the number of subplants, t_s is the duration of a time step, c_jk is the economic cost of utility j at a time step k of the prediction period, and u_jik is the rate of use of utility j by subplant i at time step k. In some embodiments, the cost function J_HL includes an additional demand charge term such as:
w_dc_demand max_nh(u_elec(θ_HL), u_max,ele)   Eq. (3)
where w_d is a weighting term, d_demand is the demand cost, and the max( ) term selects the peak electricity use during the applicable demand charge period.

In some embodiments, the high level optimization performed by the high level optimizer 440 is the same or similar to the high level optimization process described in U.S. patent application Ser. No. 14/634,609 filed Feb. 27, 2015 and titled “High Level Central Plant Optimization,” which is incorporated by reference herein.

The low level optimizer 450 receives the Q allocation data 442 from the high level optimizer 440, and determines operating parameters (e.g., capacities) of the HVAC devices of the HVAC system 100. In one or more embodiments, the low level optimizer 450 includes an equipment allocator 460, a state predictor 470, and a power estimator 480. Together, these components operate to determine a set of operating parameters, for example, rendering reduced power consumption of the HVAC system 100 for a given set of thermal energy loads indicated by the Q allocation data 442, and generate operating parameter data indicating the determined set of operating parameters. In some embodiments, the low level optimizer 450 includes different, more, or fewer components, or includes components in different arrangements than shown in FIG. 4.

In one configuration, the equipment allocator 460 receives the Q allocation data 442 from the high level optimizer 440, and generates candidate operating parameter data 462 indicating a set of candidate operating parameters of HVAC devices of the HVAC system 100. The state predictor 470 receives the candidate operating parameter data 462 and predicts thermodynamic states of the HVAC system 100 at various locations for the set of candidate operating parameters. The state predictor 470 generates state data 474 indicating the predicted thermodynamic states, and provides the state data 474 to the power estimator 480. The power estimator 480 predicts, based on the state data 474, total power consumed by the HVAC system 100 operating according to the set of candidate operating parameters, and generates the power estimation data 482 indicating the predicted power consumption. The equipment allocator 460 may repeat the process with different sets of candidate operating parameters to obtain predicted power consumptions of the HVAC system 100 operating according to different sets of candidate operating parameters, and select a set of operating parameters rendering lower power consumption. The equipment allocator 460 may generate the operating parameter and power estimation data 448 indicating (i) the selected set of operating parameters and (ii) predicted power consumption of the power plant when operating according to the selected set of operating parameters, and provide the operating parameter and power estimation data 448 to the high level optimizer 440.

Referring to FIG. 5, illustrated is a schematic representation 500 of an HVAC system, according to some embodiments. In FIG. 5, the supply device 510 (e.g., chiller) and the load device 520 (e.g., load coil) may form a loop. In this configuration, the supply device 510 supplies gas or liquid, and the load device 520 consumes the gas or liquid for controlling temperature of a space.

During periods of low load, thermal energy storages (e.g., chillers) are often cycled to meet the cooling loads of the connected buildings. A chiller may be shut off once the chilled water temperature reaches set point or a low threshold value (e.g., 40° F.) and the chiller may operate at the minimum load. The chiller then may be left off until the return water temperature reaches a high threshold value (e.g., 55° F.). Such fixed rule based system according to fixed lower limit and upper limit of temperature thresholds may be inefficient.

Instead of a fixed rule based method, the central plant controller 410 may dynamically set bounds on the chilled water temperature, determine the effective thermal mass of the water in the loop, and predict the increase or decrease in temperature based on the under or over production of chilled water dynamically. The central plant controller 410 can then keep the temperature within the bounds and use the storage to produce a behavior similar to the chiller cycling based on temperatures as well as use the additional storage for trimming the demand to reduce the demand charge.

Looking at the temperature from bulk model point of view, the bulk water temperature should follow a differential equation,
mc_p{dot over (T)}={dot over (Q)}_l−{dot over (Q)}_c   Eq. (4)
where m is the aggregate mass (or an effective thermal mass) of the water, and c_p is the specific heat capacity of water. In the form of an energy balance, this can be rearranged as:
0={dot over (Q)}_l−{dot over (Q)}_c−mc_p{dot over (T)}  Eq. (5)
0={dot over (Q)}_l−{dot over (Q)}_c−{dot over (Q)}_wss   Eq. (6)
where {dot over (Q)}_wss is the amount of “cooling discharged” from the water mass storage by allowing the temperature to increase. {dot over (Q)}_wss may be also referred to as “a deferred load” or “an induced load but not supplied.” From Eq. (6), the water mass storage acts as a standard storage element from a high level point of view.

The temperature T is meant to be an aggregate or bulk average temperature of all the water in the loop. Because the supply temperatures would change quickly when the chiller is turned on, they cannot be used in the calculation of T. Also, when the chiller and primary pumping is off, the primary return temperature will not see significant flow and should not also be used in the calculation of T. The temperature that can be used is the secondary return water temperature. This value is also filtered by the coils throughout the loop. In the case where several chillers supply the same loop, the aggregate temperature can be the weighted average of all the secondary return temperatures in the loop.

The aggregate mass of the water in the loop may not be known. Under the assumption of the water mass thermodynamics shown in Eq. (4) and with the aggregate temperature T=Tsr defined, the bulk mass or effective thermal mass m can be found using historical data of secondary return temperatures and chilled water production. In the case where building load data is available, the water mass can be estimated by approximating the derivative of the secondary return temperature using a forward finite difference and using linear regression to find the best fit of Eq. (7). The data used is that where the return water temperature is in a transient state,

$\begin{array}{cc}{\mathrm{mc}}_p\frac{\left(T_{\mathrm{sr},k}-T_{\mathrm{sr},k-1}\right)}{\Delta t}={\stackrel{.}{Q}}_l-{\stackrel{.}{Q}}_c& \mathrm{Eq}.\left(7\right)\end{array}$
where T_sr,k is the secondary return temperature at kth sample (or kth time slot).

Often, building load data may not available (no building load meters), and only historical chiller production and secondary return temperatures may be available. Transient return water temperatures may be utilized for estimating the water mass. During those periods, the chillers may be turned off. Thus, the central plant controller 410 may predict the load while the chillers are turned off.

During steady-state conditions, the contribution of the water mass storage is zero and the measured production (flow times ΔT) is equal to the load,
{dot over (Q)}_c={dot over (Q)}_l→{dot over (m)}c_p(T_ps−T_pr)={dot over (Q)}_l   Eq. (8)
where T_ps is the primary supply temperature and T_pr is the primary return temperature.

Using data from steady-state operation, the central plant controller 410 (e.g., load predictor 432) develops a predictor {dot over (Q)}_l of the load as a function of the time of day, day of week, and outside air temperature (or enthalpy).

$\begin{array}{cc}{\stackrel{\stackrel{.}{^}}{Q}}_l=f\left(T_{\mathrm{OA}},t\right)& \mathrm{Eq}.\left(9\right)\end{array}$

In one aspect, the predictor {circumflex over ({dot over (Q)})}_l of the load determined during the steady-state temperature condition may be applied to determine deferred load {dot over (Q)}_wss and thermal mass m during the non-steady state temperature condition. Instead of using data steady-state temperature conditions, the central plant controller 410 can apply a Golay filter to the production data to smooth out any transients when the temperature is changing, and determine the predictor {circumflex over ({dot over (Q)})}_l the filtered data. At times when the secondary return temperature is not at a steady-state, the chilled water production can be subtracted from the estimated load to produce an estimate of the heat flow from the water mass storage.

$\begin{array}{cc}{\mathrm{mc}}_p\frac{\left(T_{\mathrm{sr},k}-T_{\mathrm{sr},k-1}\right)}{\Delta t}=f\left(T_{\mathrm{OA}},t\right)-{\stackrel{.}{Q}}_c& \mathrm{Eq}.\left(10\right)\end{array}$
With the estimate of the load, the heat capacity mc_p can be estimated using linear regression by finding the best fit of Eq. (10) when the return water temperature is not constant.

Referring to FIG. 6, illustrated is an example timing diagram of predicting an estimated load consumption by a load device of an HVAC system, according to some embodiments. In FIG. 6, a plot 610 illustrates a thermal energy load produced by a supply device (e.g., chiller). The plot 610 may be generated based on load data from a sensor coupled to a supply device. The plot 615 illustrates a filtered result of the thermal energy load produced by the supply device. The filtered result may be a regression model of the thermal energy load produced by the supply device. In one aspect, the supply device may be disabled or turned off during time periods 618A, 618B, 618C . . . 618F. The load predictor 432 may apply filter to the load data such that a non-zero thermal energy load produced can be predicted as indicated by the plot 615 during the time periods 618A, 618B, 618C . . . 618F. The load predictor 432 may also obtain temperature of secondary return water as indicated by the plot 620. The temperature may be measured by a sensor coupled to the supply device. Based on the filtered result and the temperature of the secondary return water, the load predictor 432 may obtain an estimated load consumption by a load device as indicated by the plot 625.

The mass storage model generator 428 generates a model of mass storage (e.g., water mass storage). The estimate of the water mass and the model of the dynamics of the water mass allows generation of historical load data, even when the load is not directly measured.

Furthermore, with the mass known, the water mass of the loop can now be defined as energy storage element in the optimization problem. The energy capacity of the water mass storage can be found by taking the difference between the maximum and minimum allowable return water temperature and multiplying by the heat capacity mc_p,
C_wss=mc_p(T_sr,max−T_sr,min)   Eq. (11)
where C_wss is an energy capacity of water mass storage. The range of return water temperatures T_sr,max−T_sr,min may be predetermined or manually entered by a user.

The state of charge of the water mass storage or the charge fraction estimate is the amount of charge left in the storage element divided by the total state of charge. For the case of simple water mass storage, charge fraction may be determined as followed.

$\begin{array}{cc}\mathrm{charge}\mathrm{fraction}=Q_{\mathrm{wss}}/C_{\mathrm{wss}}=\frac{{\mathrm{mc}}_p\left(T_{\mathrm{sr},\max }-T_{\mathrm{sr},k}\right)}{{\mathrm{mc}}_p\left(T_{\mathrm{sr},\max }-T_{\mathrm{sr},\min }\right)}=\frac{\left(T_{\mathrm{sr},\max }-T_{\mathrm{sr},k}\right)}{\left(T_{\mathrm{sr},\max }-T_{\mathrm{sr},\min }\right)}& \mathrm{Eq}.\left(12\right)\end{array}$

The maximum charge and discharge rate of the water mass storage would be dependent on several factors including: the connected coils (and the resultant aggregate coil model), the current water temperature, the supply air temperature of the connected coils, etc. The maximum charge/discharge rates of the coil may be difficult to measure. In one aspect, the expected maximum and minimum change in return water temperature may be predefined or manually entered by a user through a user interface. The mass storage model generator 428 may automatically determine the charge and discharge rates based on the heat capacitance and the expected maximum and minimum change in return water temperature. Additionally, the maximum and minimum charge fractions may be ‘1’ and ‘0,’ respectively.

Given the calculated water mass storage element capacity, charge and discharge rates, and the maximum and minimum charge fractions, the central plant controller 410 can determine when to defer chiller production and when to overproduce by considering the water mass storage. The dispatched charge and discharge rates and state-of-charge over the horizon allows for the calculation of the estimate of the return water temperature over the horizon.

Examples of the design characteristics of the water mass storage include: Charge Fraction, Design Charge Rate, Design Discharge Rate, Energy Capacity, Minimum Charge Fraction, and Maximum Charge Fraction. The characteristics may be determined from the following commissionable input parameters with exemplary units: Heat Capacity, (mc_p), [kWh/degC], Maximum allowable Water Temperature, T_max, [degC], Minimum allowable Water Temperature, T_min, [degC], Maximum allowable Rate Of Water Temperature Increase, {dot over (T)}_max^↑, [degC/s], Maximum allowable Rate Of Water Temperature Decrease, {dot over (T)}_max^↓[degC/s]. The characteristics may be further determined based on the following inputs from the BMS: Secondary Return Water Temperature, T_sr, [degC].

The design characteristics can be calculated as shown in the following table:

Water Mass Storage	Chilled Water	Hot Water
Design Characteristics	Mass Storage	Mass Storage
Charge Fraction	$\frac{\left(T_{\mathrm{ma}x}-T_{\mathrm{sr}}\right)}{\left(T_{\mathrm{ma}x}-T_{min}\right)}$	$\frac{\left(T_{\mathrm{sr}}-T_{min}\right)}{\left(T_{\mathrm{ma}x}-T_{min}\right)}$
Energy Capacity	(mcp)(Tmax − Tmin)	(mcp)(Tmax − Tmin)
Design Charge Rate	(mcp){dot over (T)}max↓	(mcp){dot over (T)}max↑
Design Discharge Rate	(mcp){dot over (T)}max↑	(mcp){dot over (T)}max↓
MinimumCharge	0	0
Fraction
Maximum Charge	1	1
Fraction
Allocator (kernel)	Chilled Water	Hot Water
Output	Mass Storage	Mass Storage
Secondary Return Water	Tmax − × (Tmax − Tmin)	Tmin +  × (Tmax − Tmin)
Temperature Estimate
where  is a charge fraction.

Although the process described herein are provided with respect to water mass storage, the principles disclosed herein may be applicable to any mass storage of other liquid or gas.

FIG. 7 is a flow chart illustrating a process 700 of operating an energy plant based on a model of mass storage, according to some embodiments. The process 700 may be performed by the high level optimizer 440 of FIG. 4. In some embodiments, the process 700 may be performed by other entities. In some embodiments, the process 700 may include additional, fewer, or different steps than shown in FIG. 7.

The high level optimizer 440 obtains load data indicating a produced thermal energy load produced by a supply device in a loop (step 710). The supply device may be a chiller producing gas or liquid for consumption by a load device in the loop. The high level optimizer 440 obtains temperature data indicating temperature of gas or liquid in the loop (step 720). The high level optimizer 440 obtains weather data indicating history of weather, for example, near a building, near an energy plant, or a place, at which a climate is controlled by the energy plant (step 730).

The high level optimizer 440 generates a regression model of the produced thermal energy load produced by the supply device (step 740). The regression model may indicate a relationship between a produced load by the supply device, weather, and temperature of gas or liquid in the loop. The high level optimizer 440 may filter the produced thermal energy load to obtain the regression model. The high level optimizer 440 predicts thermal energy load production by the supply device for a training period based on the regression model (step 750).

The high level optimizer 440 determines heat capacity mc_p based on the predicted thermal energy load production (step 760). In one approach, the high level optimizer 440 generates the predictor {circumflex over ({dot over (Q)})}_l indicating a predicted load consumed by the load device, and determines the heat capacity mc_p based as described above with respect to Eq. (9) and Eq. (10). In one approach, the high level optimizer 440 determines the predictor {circumflex over ({dot over (Q)})}_l according to the produced load {dot over (Q)}_c produced by the supply device during a steady-state temperature condition, because the temperature difference is zero during the steady-state temperature condition. Moreover, the heat capacity mc_p during a non-steady state temperature condition can be determined by applying the produced load {dot over (Q)}_c determined in the steady-state temperature condition.

The high level optimizer 440 generates the model of mass storage based on the heat capacity (step 770). Characteristics of the model of mass storage may be determined based on the heat capacity. The characteristics of the model may be also determined based on a user input of a limited number of input parameters. For example, charge fraction, design charge rate, design discharge rate, energy capacity may be determined based on the heat capacity, maximum allowable water temperature, T_max, minimum allowable water temperature, T_min, maximum allowable rate of water Temperature Increase, {dot over (T)}_max^↑, and maximum allowable rate of water temperature decrease, {dot over (T)}_max^↓.

The high level optimizer 440 predicts induced load of the load device based on the model of mass storage (step 780). For example, the high level optimizer 440 can determine when to defer chiller production and when to overproduce by considering the water mass storage. In one aspect, the high level optimizer 440 may obtain a temperature measurement and a produced thermal energy load for a time period, and apply the temperature measurement and a produced thermal energy load to the Eq. (10) to predict the induced load. In another aspect, the high level optimizer 440 may obtain a predicted temperature measurement and a predicted produced thermal energy load in the future, and apply the predicted temperature measurement and the predicted produced thermal energy load to the Eq. (10) to predict the induced load in the future. The high level optimizer 440 may also predict the induced load of the load device based on a weather forecast.

FIG. 8 is a flow chart illustrating another process 800 of operating an energy plant, according to some embodiments. The process 800 may be performed by the high level optimizer 440 of FIG. 4. In some embodiments, the process 800 may be performed by other entities. In some embodiments, the process 800 may include additional, fewer, or different steps than shown in FIG. 8.

The high level optimizer 440 obtains a maximum allowable temperature and a minimum allowable temperature of gas or liquid in a loop (step 810). The maximum allowable temperature and the minimum allowable temperature of the gas or the liquid may be obtained by a user through a user interface. Alternatively, the maximum allowable temperature and the minimum allowable temperature of the gas or the liquid may be predetermined.

The high level optimizer 440 generates a model indicating how a difference between induced load at the load device and produced load by a supply device affects a temperature of the gas or the liquid in the loop (step 820). For example, the high level optimizer obtains the model as the secondary return water temperature estimate based on the charge fraction SŌC.

The high level optimizer 440 generates a cost function with a constraint to conform to the model (step 830). The constraints may be to maintain the temperature to be within a predetermined range set by the maximum allowable temperature and the minimum allowable temperature.

The high level optimizer 440 determines control decision values based on the cost function (step 840). For example, the high level optimizer 440 determines control decision values by minimizing the cost function while satisfying the constraint. Examples of the control decision values include when to defer the produced load by the supply device and an amount of the produced load. The high level optimizer 440 controls the energy plant according to the control decision values (step 850).

In some embodiments, the high level optimizer 440 obtains load data indicating thermal energy load produced. The load data may be obtained by a sensor coupled to a supply device supplying gas or liquid to a load device in a loop.

In some embodiments the high level optimizer 440 predicts thermal energy load consumption for a first time period by a load device based on the load data. The supply device may be disabled or turned off during a time period within the first time period. In one approach, the high level optimizer 440 applies filtering (e.g., Golay filtering) on the thermal energy produced. The filtered result may render a non-zero thermal energy load produced during the time period. The high level optimizer 440 may obtain temperature data indicating return temperature (e.g., secondary return temperature). Based on the filtered result and the returned temperature, the high level optimizer 440 may predict thermal energy load consumption for the first time period. In some embodiments, the high level optimizer 440 determines the time period, during which supply device is turned off, and applies filtering on the thermal energy produced for the time period.

In some embodiments, the high level optimizer 440 generates a model of mass storage (e.g., water mass storage) based on the predicted thermal energy load consumption. The high level optimizer 440 may determine a heat capacity of gas or liquid in the loop based on the predicted thermal energy load consumption, for example, according to Eq. (10). Based on the heat capacity, the high level optimizer 440 may determine other characteristics of the model of mass storage. Examples of characteristics of the model of mass storage include a charge rate, a discharge rate, and an energy capacity. The high level optimizer 440 may also determine the characteristics of the model of mass storage based on a few number of input parameters. Examples of the parameters include a maximum allowable temperature of the gas or the liquid, a minimum allowable temperature of the gas or the liquid, a maximum allowable rate of increase in temperature of the gas or the liquid, and a minimum allowable rate of increase in temperature of the gas or the liquid. The parameters may be predefined, and/or the high level optimizer 440 may obtain the parameters from a user through a user interface. Based on the characteristics of the model of mass storage, the high level optimizer 440 may automatically generate the model of mass storage.

In some embodiments, the high level optimizer 440 determines an amount of productions of gas or liquid by a supply device based on the model of mass storage. The high level optimizer 440 may determine an amount of production of the gas or the liquid by the supply device for a second time period according to the model of the mass storage. The second time period may be after the first time period.

In some embodiments, the high level optimizer 440 operates the energy plant according to the determined amount of production of gas or liquid.

In some embodiments, the high level optimizer 440 determines an effective thermal mass in a loop. The high level optimizer 440 may obtain load data indicating thermal energy load produced. The high level optimizer 440 may predict thermal energy load consumption for a first time period by a load device based on the load data. The high level optimizer 440 may apply filtering (e.g., Golay filtering) on the thermal energy produced. By filtering the thermal energy load produced, a non-zero thermal energy load produced when the supply device is turned off in the first time period can be predicted. Moreover, the high level optimizer 440 can predict a non-zero thermal energy load consumption of the load device when the supply device is turned off based on the non-zero thermal energy load produced. The high level optimizer 440 may also obtain temperature data indicating return temperature (e.g., secondary return temperature). Based on the filtered result and the returned temperature, the high level optimizer 440 may predict thermal energy load consumption for the first time period. The high level optimizer 440 may determine the effective thermal mass based on the predicted thermal energy load consumption, for example, according to Eq. (10).

In some embodiments, the high level optimizer 440 obtains characteristics of a water mass storage based on the effective thermal mass. The high level optimizer 440 may determine a charge rate, a discharge rate, and an energy capacity of a model of mass storage in the loop during the first time period based on the effective thermal energy mass.

In some embodiments, the high level optimizer 440 predicts a change in temperature of gas or liquid in the loop. In one approach, the high level optimizer 440 predicts an amount of production of gas or liquid in the loop during a second time period based on the effective thermal mass. The second time period may be after the first time period. The high level optimizer 440 may predict a change in temperature of gas or liquid in the loop based on the predicted amount of production of gas or liquid in the loop.

In some embodiments, the high level optimizer 440 adjusts thermal energy load consumed by a load device according to the predicted change in the temperature of gas or liquid. For example, the high level optimizer 440 controls the temperature of the gas or the liquid in the loop to be within an allowable temperature range during the second time period.

Configuration of Exemplary Embodiments

The construction and arrangement of the systems and methods as shown in the various exemplary embodiments are illustrative only. Although only a few embodiments have been described in detail in this disclosure, many modifications are possible (e.g., variations in sizes, dimensions, structures, shapes and proportions of the various elements, values of parameters, mounting arrangements, use of materials, colors, orientations, etc.). For example, the position of elements may be reversed or otherwise varied and the nature or number of discrete elements or positions may be altered or varied. Accordingly, all such modifications are intended to be included within the scope of the present disclosure. The order or sequence of any process or method steps may be varied or re-sequenced according to alternative embodiments. Other substitutions, modifications, changes, and omissions may be made in the design, operating conditions and arrangement of the exemplary embodiments without departing from the scope of the present disclosure.

The present disclosure contemplates methods, systems and program products on any machine-readable media for accomplishing various operations. The embodiments of the present disclosure may be implemented using existing computer processors, or by a special purpose computer processor for an appropriate system, incorporated for this or another purpose, or by a hardwired system. Embodiments within the scope of the present disclosure include program products comprising machine-readable media for carrying or having machine-executable instructions or data structures stored thereon. Such machine-readable media can be any available media that can be accessed by a general purpose or special purpose computer or other machine with a processor. By way of example, such machine-readable media can include RAM, ROM, EPROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other medium which can be used to carry or store desired program code in the form of machine-executable instructions or data structures and which can be accessed by a general purpose or special purpose computer or other machine with a processor. Combinations of the above are also included within the scope of machine-readable media. Machine-executable instructions include, for example, instructions and data which cause a general purpose computer, special purpose computer, or special purpose processing machines to perform a certain function or group of functions.

Although the figures show a specific order of method steps, the order of the steps may differ from what is depicted. Also two or more steps may be performed concurrently or with partial concurrence. Such variation will depend on the software and hardware systems chosen and on designer choice. All such variations are within the scope of the disclosure. Likewise, software implementations could be accomplished with standard programming techniques with rule based logic and other logic to accomplish the various connection steps, processing steps, comparison steps and decision steps.

Claims

1. A controller for an energy plant including a fluid loop formed by a plurality of devices, the controller comprising:

a processing circuit comprising a processor and memory storing instructions executed by the processor, the processing circuit configured to:

obtain a maximum allowable temperature and a minimum allowable temperature of gas or liquid in the fluid loop;

generate a model indicating a relationship between (i) a temperature of the gas or the liquid in the fluid loop, and (ii) a difference between a first thermal energy load removed from the fluid loop by a load device of the plurality of devices and a second thermal energy load supplied to the fluid loop by a supply device of the plurality of devices;

generate a cost function with a constraint according to the model;

determine control decision values based on the cost function; and

operate the energy plant according to the control decision values to control the temperature of the gas or liquid in the fluid loop by adjusting the second thermal energy load supplied to the fluid loop by the supply device.

2. The controller of claim 1, wherein the constraint is to keep the temperature of the gas or the liquid in the fluid loop to be between the maximum allowable temperature and the minimum allowable temperature, when the energy plant operates according to the control decision values.

3. The controller of claim 1, wherein the control decision values include when to defer the second thermal energy load supplied to the fluid loop by the supply device and an amount of the second thermal energy load.

4. The controller of claim 1, wherein the cost function corresponds to a total energy consumed by the energy plant.

5. The controller of claim 1, wherein the control decision values are determined to minimize the cost function, while complying with the constraint.

6. The controller of claim 1, wherein the processing circuit is configured to:

generate a regression model of the second thermal energy load;

predict the second thermal energy load supplied to the fluid loop by the supply device for a first time period based on the regression model;

determine a heat capacity of the gas or the liquid in the fluid loop based on the predicted second thermal energy load; and

generate the model based on the heat capacity.

7. The controller of claim 6, wherein the processing circuit is configured to:

obtain load data indicating the second thermal energy load during a second time period, the second time period before the first time period,

wherein the processing circuit is configured to generate the regression model based on the load data.

8. The controller of claim 7, wherein the processing circuit is configured to:

filter the second thermal energy load during the second time period,

wherein the processing circuit is configured to generate the regression model based on the filtered thermal energy load.

9. A method for an energy plant including a fluid loop formed by a plurality of devices, the method including:

obtaining a maximum allowable temperature and a minimum allowable temperature of gas or liquid in the fluid loop;

generating a model indicating a relationship between (i) a temperature of the gas or the liquid in the fluid loop, and (ii) a difference between a first thermal energy load removed from the fluid loop by a load device of the plurality of devices and a second thermal energy load supplied to the fluid loop by a supply device of the plurality of devices;

generating a cost function with a constraint according to the model;

determining control decision values based on the cost function; and

operating the energy plant according to the control decision values to control the temperature of the gas or liquid in the fluid loop by adjusting the second thermal energy load supplied to the fluid loop by the supply device.

10. The method of claim 9, wherein the constraint is to keep the temperature of the gas or the liquid in the fluid loop to be between the maximum allowable temperature and the minimum allowable temperature, when the energy plant operates according to the control decision values.

11. The method of claim 9, wherein the control decision values include when to defer the second thermal energy load supplied to the fluid loop by the supply device and an amount of the second thermal energy load.

12. The method of claim 9, wherein the cost function corresponds to a total energy consumed by the energy plant.

13. The method of claim 9, wherein the control decision values are determined to minimize the cost function, while complying with the constraint.

14. The method of claim 9, further comprising:

generating a regression model of the second thermal energy load;

predicting the second thermal energy load supplied to the fluid loop by the supply device for a first time period based on the regression model;

determining a heat capacity of the gas or the liquid in the fluid loop based on the predicted second thermal energy load; and

generate the model based on the heat capacity.

15. The method of claim 14, further comprising:

obtaining load data indicating the second thermal energy load during a second time period by the supply device, the second time period before the first time period,

wherein the regression model is generated based on the load data.

16. The method of claim 15, further comprising:

filtering the second thermal energy load during the second time period,

wherein the regression model is generated based on the filtered thermal energy load.

17. A non-transitory computer readable medium storing instructions for an energy plant including a fluid loop formed by a plurality of devices, the instructions when executed by a processor cause the processor to:

obtain a maximum allowable temperature and a minimum allowable temperature of gas or liquid in the fluid loop;

generate a model indicating a relationship between (i) a temperature of the gas or the liquid in the fluid loop, and (ii) a difference between a first thermal energy load removed from the fluid loop by a load device of the plurality of devices and a second thermal energy load supplied to the fluid loop by a supply device of the plurality of devices;

generate a cost function with a constraint according to the model;

determine control decision values based on the cost function; and

operate the energy plant according to the control decision values to control the temperature of the gas or liquid in the fluid loop by adjusting the second thermal energy load supplied to the fluid loop by the supply device.

18. The non-transitory computer readable medium of claim 17, wherein the constraint is to keep the temperature of the gas or the liquid in the fluid loop to be between the maximum allowable temperature and the minimum allowable temperature, when the energy plant operates according to the control decision values.

19. The non-transitory computer readable medium of claim 17, wherein the control decision values include when to defer the second thermal energy load supplied to the fluid loop by the supply device and an amount of the second thermal energy load.

20. The non-transitory computer readable medium of claim 17, wherein the cost function corresponds to a total energy consumed by the energy plant, and wherein the control decision values are determined to minimize the cost function, while complying with the constraint.

21. A controller for a fluid loop formed by a plurality of devices, the controller comprising:

a processing circuit comprising a processor and memory storing instructions executed by the processor, the processing circuit configured to:

obtain a maximum allowable temperature and a minimum allowable temperature of gas or liquid in the fluid loop;

generate a model indicating a relationship between (i) a temperature of gas or liquid in the fluid loop and (ii) a difference between a first thermal energy load removed from the fluid loop by a load device of the plurality of devices and a second thermal energy load supplied to the fluid loop by a supply device of the plurality of devices;

generate a cost function with a constraint according to the model;

determine control decision values based on the cost function; and

operate the supply device according to the control decision values to control the temperature of the gas or liquid in the fluid loop by adjusting the second thermal energy load supplied to the fluid loop by the supply device.

22. The controller of claim 21, wherein the control decision values comprise a deferred thermal energy load indicating an amount of thermal energy to discharge from the fluid loop in excess of the second thermal energy load supplied to the fluid loop by the supply device.

23. The controller of claim 21, wherein the control decision values comprise a deferred thermal energy load which causes the temperature of the gas or liquid in the fluid loop to increase toward the maximum allowable temperature or decrease toward the minimum allowable temperature when the deferred thermal energy load is discharged from the fluid loop.

24. The controller of claim 21, wherein the control decision values comprise when to defer the second thermal energy load supplied to the fluid loop by the supply device based on a thermal mass of the gas or liquid in the fluid loop.

25. The controller of claim 21, wherein the processing circuit is configured to predict an increase or decrease in the temperature of the gas or liquid in the fluid loop based on an amount of underproduction or overproduction of the second thermal energy load supplied to the fluid loop by the supply device.

26. The controller of claim 21, wherein the processing circuit is configured to determine a thermal mass of the gas or liquid in the fluid loop using the model.

27. The controller of claim 26, wherein the processing circuit is configured to determine an energy capacity of the gas or liquid in the fluid loop based on the thermal mass of the gas or liquid in the fluid loop.

28. A method for operating a fluid loop formed by a plurality of devices, the method comprising:

obtaining a maximum allowable temperature and a minimum allowable temperature of gas or liquid in the fluid loop;

generating a model indicating a relationship between (i) a temperature of gas or liquid in the fluid loop and (ii) a difference between a first thermal energy load removed from the fluid loop by a load device of the plurality of devices and a second thermal energy load supplied to the fluid loop by a supply device of the plurality of devices;

generating a cost function with a constraint according to the model;

determining control decision values based on the cost function; and

operating the supply device according to the control decision values to control the temperature of the gas or liquid in the fluid loop by adjusting the second thermal energy load supplied to the fluid loop by the supply device.

29. The method of claim 28, wherein the control decision values comprise a deferred thermal energy load indicating an amount of thermal energy to discharge from the fluid loop in excess of the second thermal energy load supplied to the fluid loop by the supply device.

30. The method of claim 28, wherein the control decision values comprise a deferred thermal energy load which causes the temperature of the gas or liquid in the fluid loop to increase toward the maximum allowable temperature or decrease toward the minimum allowable temperature when the deferred thermal energy load is discharged from the fluid loop.

31. The method of claim 28, wherein the control decision values comprise when to defer the second thermal energy load supplied to the fluid loop by the supply device based on a thermal mass of the gas or liquid in the fluid loop.

32. The method of claim 28, comprising predicting an increase or decrease in the temperature of the gas or liquid in the fluid loop based on an amount of underproduction or overproduction of the second thermal energy load supplied to the fluid loop by the supply device.

33. The method of claim 28, comprising determining a thermal mass of the gas or liquid in the fluid loop using the model.

34. The method of claim 33, comprising determining an energy capacity of the gas or liquid in the fluid loop based on the thermal mass of the gas or liquid in the fluid loop.

35. One or more non-transitory computer readable media storing instructions for operating a fluid loop formed by a plurality of devices, wherein the instructions, when executed by one or more processors, cause the one or more processors to:

obtain a maximum allowable temperature and a minimum allowable temperature of gas or liquid in the fluid loop;

generate a model indicating a relationship between (i) a temperature of gas or liquid in the fluid loop and (ii) a difference between a first thermal energy load removed from the fluid loop by a load device of the plurality of devices and a second thermal energy load supplied to the fluid loop by a supply device of the plurality of devices;

generate a cost function with a constraint according to the model;

determine control decision values based on the cost function; and

operate the supply device according to the control decision values to control the temperature of the gas or liquid in the fluid loop by adjusting the second thermal energy load supplied to the fluid loop by the supply device.

36. The non-transitory computer readable media of claim 35, wherein the control decision values comprise a deferred thermal energy load indicating an amount of thermal energy to discharge from the fluid loop in excess of the second thermal energy load supplied to the fluid loop by the supply device.

37. The non-transitory computer readable media of claim 35, wherein the control decision values comprise a deferred thermal energy load which causes the temperature of the gas or liquid in the fluid loop to increase toward the maximum allowable temperature or decrease toward the minimum allowable temperature when the deferred thermal energy load is discharged from the fluid loop.

38. The non-transitory computer readable media of claim 35, wherein the control decision values comprise when to defer the second thermal energy load supplied to the fluid loop by the supply device based on a thermal mass of the gas or liquid in the fluid loop.

39. The non-transitory computer readable media of claim 35, wherein the instructions cause the one or more processors to predict an increase or decrease in the temperature of the gas or liquid in the fluid loop based on an amount of underproduction or overproduction of the second thermal energy load supplied to the fluid loop by the supply device.

40. The non-transitory computer readable media of claim 35, wherein the instructions cause the one or more processors to determine at least one of a thermal mass of the gas or liquid in the fluid loop or an energy capacity of the gas or liquid in the fluid loop using the model.