A chiller control method, device, equipment and storage medium
By presetting the opening and closing actions and quantity of chiller units, combining meteorological and historical data to predict future cooling loads, the target number is screened to control the opening and closing of chiller units, and solving the energy waste and reliability caused by frequent opening and closing of chiller units in the existing technology in the existing technology Problem, improving system stability and energy efficiency.
Patent Information
- Application Number
- CN202410250091.8
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- Filing Date
- 2024-03-05
- Publication Date
- 2024-11-12
- Estimated Expiration
- 2044-03-05
AI Technical Summary
The existing chiller control method does not consider future cooling load conversion, resulting in frequent opening and closing operations, reducing system reliability and causing energy waste.
By presetting the opening and closing actions and quantity of chiller units, predicting the probability of future cooling load based on meteorological data and historical data, the target number is selected to control the opening and closing operations of chiller units, and ensuring that the chiller unit can carry cooling load in the future moments.
It reduces unnecessary opening and closing operation frequency of chiller units, improves system stability and energy utilization efficiency, and extends the service life of chiller units.
Smart Images

Figure CN118189539B_ABST
Abstract
Description
Technical Field
[0001] The present invention relates to the technical field of chillers, and in particular to a chiller control method, device, equipment and storage medium. Background Art
[0002] The chiller provides cooling to objects that need cooling, such as buildings and power equipment, through heat exchange. That is, the chiller is used to meet the cooling needs of objects that need cooling. For example, when power equipment needs chilled water for cooling, the power equipment first delivers its own higher temperature water (the cooling load that the chiller needs to bear) to the chiller. The chiller performs heat exchange on this part of water to lower the water temperature and then forms chilled water. The cooling load applied to the chiller by the object that needs cooling is not fixed over time. In order to enable the chiller to bear the above cooling load, the cooling load can be collected in real time, and the switch of the chiller can be adjusted according to the real-time cooling load. That is, when the real-time cooling load is large, more chillers are started, and when the real-time cooling load is small, more chillers are closed. Since only the real-time cooling load is considered without considering the possible cooling load changes in the future, the operation of starting and closing the chiller is performed unnecessarily frequently.
[0003] The existing technology is based on the following principle to control the opening and closing action of the chiller. The control process of the opening and closing action is as follows: Fig. 9 shown.
[0004]
[0005]
[0006]
[0007]
[0008] c w is the specific heat capacity of water (kJ / kg·℃), is the cold water flow rate in the main pipeline (kg / s), T rtn ,T sup are the return water temperature and supply water temperature of the chilled water. When the chillers have the same capacity, is the threshold for shutting down a chiller. is the threshold for starting a chiller. n is the number of chillers currently in operation, d is the cold load dead zone set by the operation and maintenance personnel, Q c is the cooling capacity of a chiller. The basic rule of chiller sequence control is that the cooling capacity provided by the chiller should meet the actual cooling load, Q t It is the cooling load measured in real time.
[0009] In summary, the prior art reduces the reliability of system operation and causes waste of energy due to unnecessary execution of the operation of starting and closing the chiller.
[0010] Therefore, the prior art still needs to be improved and enhanced. Summary of the invention
[0011] In order to solve the above technical problems, the present invention provides a chiller control method, device, equipment and storage medium, which solves the problem of energy waste caused by unnecessary execution of chiller start and stop operations in the prior art.
[0012] To achieve the above object, the present invention adopts the following technical solutions:
[0013] In a first aspect, the present invention provides a chiller control method, comprising:
[0014] Determine a preselected action of the chiller and a preset number of the chillers that need to perform the preselected action, the action being used to represent the action of starting and closing the chiller;
[0015] After the preset number of the chillers perform the preselected actions, the limit cooling load that the chillers can bear is determined;
[0016] Acquire meteorological data at the current moment, and based on the meteorological data, predict the predicted probability that the extreme cooling load matches the future cooling load, where the future cooling load is the cooling load that the chiller needs to bear at a future moment;
[0017] According to the predicted probability, a target number is screened out from the preset number, and at the future moment, the target number of chillers is controlled to perform the preset action.
[0018] In one implementation, the determining of a preselected action of a chiller and a preset number of chillers that need to perform the preselected action includes:
[0019] Obtaining the current cooling load carried by the chiller at the current moment, the number of the chillers that have been started and are in the started state at the current moment, and the maximum cooling load that each chiller can carry, and determining a preliminary selection action and an initial number of the chillers that need to perform the preliminary selection action based on the current cooling load, the number of the chillers that have been started and the maximum cooling load;
[0020] using the primary selection action as the pre-selected action;
[0021] The set quantity is subtracted from the initial quantity to obtain an alternative quantity, and the initial quantity and the alternative quantity are used as the preset quantity.
[0022] In one implementation, the determining of the preliminary action and the initial number of the chillers that need to perform the preliminary action based on the current cooling load, the started number and the maximum cooling load includes:
[0023] The first intermediate formula is obtained by adding the started number to the set parameter and subtracting 1, and the first intermediate formula is multiplied by the maximum cooling load to construct a first formula;
[0024] Adding the started quantity to the parameter to obtain a second intermediate formula, and multiplying the second intermediate formula by the maximum cooling load to construct a second formula;
[0025] The first formula is set to be smaller than the current cooling load, and the current cooling load is smaller than or equal to the second formula, so as to construct an inequality;
[0026] An integer of the parameter satisfying the inequality is determined, and the absolute value of the integer is used as the initial quantity.
[0027] In one implementation, after the preset number of chillers perform the preselected action, determining the limit cooling load that the chillers can bear includes:
[0028] If the preselected action is a shut-down action, the result of subtracting the initial set number from the started number is multiplied by the maximum cooling load to obtain the minimum cooling load that the chiller can bear;
[0029] The maximum cooling load that the chiller can bear is obtained by subtracting the alternative number from the started number and multiplying it by the maximum cooling load, and the maximum cooling load and the minimum cooling load are the limit cooling load.
[0030] In one implementation, the predicting, based on the meteorological data, the predicted probability that the extreme cooling load matches the future cooling load includes:
[0031] Obtaining calendar information corresponding to the current moment, and obtaining historical cooling load data of the chiller;
[0032] Applying a natural gradient boosting algorithm to the meteorological data, the calendar information, and the historical cooling load data to predict the probability distribution of the future cooling load;
[0033] According to the probability distribution, a first predicted probability that the future cooling load is less than the minimum cooling load and a second predicted probability that the future cooling load is less than the maximum cooling load are determined, and the first predicted probability and the second predicted probability are used as the predicted probability.
[0034] In one implementation, the step of selecting a target quantity from the preset quantity based on the predicted probability includes:
[0035] Subtract the first predicted probability from 1 to obtain a first risk value corresponding to the initial quantity;
[0036] Subtract the second predicted probability from the result of adding 1 to the first predicted probability to obtain a second risk value corresponding to the number of alternatives;
[0037] A target quantity is screened out from the initial quantity and the alternative quantities according to the first risk value and the second risk value.
[0038] In one implementation, the step of selecting a target quantity from the initial quantity and the candidate quantities based on the first risk value and the second risk value includes:
[0039] Accumulating the first risk values generated by using the initial set quantity at each of the future moments to obtain a first risk accumulated value;
[0040] Accumulating the second risk values generated by using the alternative quantities at each of the future moments to obtain a second risk accumulated value;
[0041] When the first risk cumulative value is less than the second risk cumulative value, the initial quantity is used as the target quantity;
[0042] Or, when the second risk cumulative value is less than the first risk cumulative value, the alternative quantity is used as the target quantity.
[0043] In a second aspect, an embodiment of the present invention further provides a chiller control device, wherein the control device comprises the following components:
[0044] A quantity construction module, used to determine a preselected action of a chiller and a preset number of the chillers that need to perform the preselected action, wherein the action is used to represent the action of starting and closing the chiller;
[0045] A cooling load calculation module, used for determining the ultimate cooling load that the chillers can bear after the preset number of chillers perform the preselected actions;
[0046] A probability calculation module, used to obtain meteorological data at the current moment, and based on the meteorological data, predict the predicted probability that the extreme cooling load matches the future cooling load, where the future cooling load is the cooling load that the chiller needs to bear at a future moment;
[0047] The quantity screening module is used to screen out a target quantity from the preset quantity according to the predicted probability, and control the target quantity of chillers to perform the preset action at the future moment.
[0048] In a third aspect, an embodiment of the present invention further provides a terminal device, wherein the terminal device includes a memory, a processor, and a chiller control program stored in the memory and executable on the processor, and when the processor executes the chiller control program, the steps of the above-mentioned chiller control method are implemented.
[0049] In a fourth aspect, an embodiment of the present invention further provides a computer-readable storage medium, on which a chiller control program is stored. When the chiller control program is executed by a processor, the steps of the above-mentioned chiller control method are implemented.
[0050] Beneficial effect: The present invention first presets the preselected action of opening or closing for the chiller, and presets how many chillers need to perform the preselected action, that is, determines the preset number. And it is assumed that if a preset number of chillers perform the preselected action, then the maximum cooling load that these preset number of chillers can bear together. At the same time, the predicted probability that the extreme cooling load corresponding to the preset number matches the future cooling load is predicted, and according to the predicted probability, the target number is screened out from multiple preset numbers, and the target number of chillers is controlled to perform the preselected action. From the above analysis, it can be seen that the present invention controls the number of chillers that perform the opening and closing operations by predicting the probability of whether the extreme cooling load matches the future cooling load, thereby reducing the operating frequency of the opening and closing chiller, thereby improving the stability of the chiller and saving energy. BRIEF DESCRIPTION OF THE DRAWINGS
[0051] Figure 1 It is the overall flow chart of the present invention;
[0052] Figure 2 It is a schematic diagram of the structure of a water chiller in an embodiment of the present invention;
[0053] Figure 3 This is a flow chart of the start and stop control of the chiller in an embodiment of the present invention;
[0054] Figure 4 Schematic diagram of normal probability distribution in an embodiment of the present invention;
[0055] Figure 5 A schematic diagram of a first risk value of a closing action in an embodiment of the present invention;
[0056] Figure 6 A schematic diagram of a second risk value of a closing action in an embodiment of the present invention;
[0057] Figure 7 A schematic diagram of risk values corresponding to starting a select number of chillers in an embodiment of the present invention;
[0058] Figure 8 A schematic diagram of the risk values corresponding to starting the initially set number of chillers in an embodiment of the present invention;
[0059] Fig. 9 It is a schematic diagram of the opening and closing action in the background technology;
[0060] Fig.10 A structural diagram of a chiller control device provided in an embodiment of the present invention;
[0061] Fig.11 This is a block diagram of the internal structure principle of the terminal device provided in an embodiment of the present invention. DETAILED DESCRIPTION
[0062] The following is a clear and complete description of the technical solution of the present invention in combination with the embodiments and the accompanying drawings. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0063] According to research, chillers provide cooling to objects that need cooling, such as buildings and power equipment, through heat exchange. That is, chillers are used to meet the cooling needs of objects that need cooling. For example, when power equipment needs chilled water for cooling, the power equipment first delivers its own higher temperature water (the cooling load that the chiller needs to bear) to the chiller. The chiller performs heat exchange on this part of water to lower the water temperature and then forms chilled water. The cooling load applied to the chiller by the object that needs cooling is not fixed over time. In order to enable the chiller to bear the above cooling load, the cooling load can be collected in real time, and the switch of the chiller can be adjusted according to the real-time cooling load. That is, when the real-time cooling load is large, more chillers are started, and when the real-time cooling load is small, more chillers are closed. Since only the real-time cooling load is considered without considering the possible cooling load changes in the future, the operation of starting and closing the chiller is performed unnecessarily frequently.
[0064] In order to solve the above technical problems, the present invention provides a chiller control method, device, equipment and storage medium, which solves the problem of energy waste caused by unnecessary execution of the operation of starting and closing the chiller in the prior art. In specific implementation, first determine the preselected action of the chiller and the preset number of the chillers that need to execute the preselected action, the action is used to characterize the action of starting and closing the chiller; assume that after the preset number of chillers execute the preselected action, estimate the extreme cold load that the chiller can bear; obtain the meteorological data at the current moment, and based on the meteorological data, predict the predicted probability of the extreme cold load matching the future cold load, the future cold load is the cold load that the chiller needs to bear at the future moment; finally, based on the predicted probability, select the target number from the preset number, and control the target number of chillers to execute the preset action at the future moment.
[0065] For example, if there are 10 chillers used to provide chilled water to the building, the structure of one chiller is as follows: Figure 2 As shown, after the chilled water enters the building, the ambient temperature of the building causes the chilled water to heat up to form water to be chilled with a higher temperature (this part of the water to be chilled is the cooling load that the chiller needs to bear). The chilled water enters the chiller, and the chiller cools the water to be chilled to form chilled water. This application uses the following technical means to control the switches of these 10 chillers:
[0066] For example, if it is known that 7 chillers have been started at the current moment, the maximum cooling load that each chiller can bear, and the current cooling load that the chiller needs to bear at the current moment, based on these three factors, determine how many chillers together can bear the current cooling load. For example, if it is determined that 5 chillers can bear the current cooling load, then it is necessary to shut down 2 more chillers, that is, the pre-selected action is to shut down, and the preset number can be 2 and 1. Shut down 2 units, there are 5 units left, calculate the maximum cooling load that 5 units can bear together (recorded as the minimum cooling load); shut down 1 unit, there are 6 units left, calculate the maximum cooling load that 6 units can bear together (recorded as the maximum cooling load). If the probability of predicting that the cooling load (future cooling load) in the next hour is Q1 is f1, the probability of predicting that the cooling load (future cooling load) in the next hour is Q2 is f2, and the probability of predicting that the cooling load (future cooling load) in the next hour is Q3 is f3. If among Q1, Q2, and Q3, Q2 is less than the minimum cooling load (shutting down 2 units corresponds to the minimum cooling load), and Q3 is less than the maximum cooling load (shutting down 1 unit corresponds to the maximum cooling load), then the risk 1-f2 corresponding to shutting down 2 units is calculated, and the risk f2+1-f3 corresponding to shutting down 1 unit is calculated. If 1-f2 is less than f2+1-f3, then the preset number 2 is used as the target number.
[0067] The chiller control method of this embodiment can be applied to a terminal device, which can be a terminal product with a data processing function, such as a controller. Figure 1 As shown in , the chiller control method specifically includes the following steps:
[0068] S100, determining a preselected action of a chiller and a preset number of the chillers that need to perform the preselected action, wherein the action is used to represent the action of starting and closing the chiller.
[0069] S200, after the preset number of chillers perform the preselected action, determine the ultimate cooling load that the chillers can bear.
[0070] S300, obtaining meteorological data at the current moment, and based on the meteorological data, predicting the predicted probability of the extreme cooling load matching the future cooling load, where the future cooling load is the cooling load that the chiller needs to bear at a future moment.
[0071] S400, selecting a target number from the preset number according to the predicted probability, and controlling the target number of chillers to perform the preset action at the future moment.
[0072] In one embodiment, step S100 includes the following specific steps S101 and S102:
[0073] S101, obtaining the current cooling load carried by the chiller at the current moment,
[0074] The number of started chillers that are in the started state at the moment, the maximum cooling load that each chiller can bear, and based on the current cooling load, the started number and the maximum cooling load, determine the preliminary action and the initial number of chillers that need to perform the preliminary action, and use the preliminary action as the pre-selected action.
[0075] In this embodiment, each chiller has four 650RT (2286kW) chillers.
[0076] In this embodiment, the specific process of determining the initial number is as follows: add the started number plus the set parameter minus 1 to obtain a first intermediate formula, and multiply the first intermediate formula by the maximum cooling load to construct a first formula; add the started number plus the parameter to obtain a second intermediate formula, and multiply the second intermediate formula by the maximum cooling load to construct a second formula; let the constructed first formula be less than the current cooling load, and the current cooling load be less than or equal to the second formula to construct an inequality; determine an integer of the parameter that satisfies the inequality, and use the absolute value of the integer as the initial number.
[0077] The inequality is as follows:
[0078] (n+a1-1)·Q c <Q t ≤(n+a1)·Q c
[0079] In the formula, n is the number of started, a1 is the parameter, Q c is the maximum cooling load, Q t is the current cooling load, t is the current time, n+a1-1 is the first intermediate formula, (n+a1-1)·Q c is the first equation, n+a1 is the second
[0080] Second intermediate formula, (n+a1)·Q c is the second formula. Calculate the value of a1 that satisfies the above inequality. The absolute value of a1 is the initial quantity.
[0081] As shown in Table 1, when the value of a1 is less than zero, the primary action is to shut down the chillers based on the number of chillers that have been started. Further, if the value of a1 is -2 or -1, one or two chillers are shut down, that is, after shutting down one or two chillers, the total cooling capacity provided by the remaining chillers can still meet the cooling demand of the current cooling load.
[0082] When the value of a1 is greater than zero, the primary action is to start the chillers based on the number of chillers that have been started. If the value of a1 is 2 or 1 (such as Figure 3 As shown in Figure 1, then start one or two chillers. That is, you need to start one or two more chillers, and the total cooling capacity provided by these chillers can meet the cooling demand of the current cooling load.
[0083] When the value of a1 is equal to zero, it indicates that the number of chillers that have been started is sufficient to carry the current cooling load without switching any chillers on or off, that is, the number of chillers that have been started is sufficient to cool the current cooling load.
[0084] Table 1
[0085]
[0086] S102, subtracting the set quantity from the initial quantity to obtain an alternative quantity A2, and using the initial quantity A1 and the alternative quantity A2 as the preset quantity.
[0087] In this embodiment, the set number is 1, A1 = |a1|, A2 = |a2|, A2 = A1-1.
[0088] In one embodiment, step S200 includes the following specific steps S201 and S202:
[0089] S201, if the preselected action is a shutdown action, multiply the result of subtracting the initial number A1 from the started number n by the maximum cooling load Q c , get the minimum cooling load Q that the chiller can bear l .
[0090] Q l =(n-A1)·Q c
[0091] S202: multiply the result of subtracting the candidate quantity A2 from the started quantity n by the maximum cooling load Q c , get the maximum cooling load Q that the chiller can bear h , and the maximum cooling load and the minimum cooling load are the limit cooling load. h Satisfies the following relationship:
[0092] Q h =(n-A2)·Q c
[0093] In this embodiment, if the preselected action is the opening action, the minimum cooling load is recorded as Q' l , Q′ l =(n+A2)·Q c , the maximum cooling load is recorded as Q' h , Q′ h =(n+A1)·Q c .
[0094] In one embodiment, step S300 includes the following specific steps S301, S302, and S303:
[0095] S301, obtaining calendar information corresponding to the current moment, and obtaining historical cooling load data of the chiller.
[0096] Calendar information is calendar features, including whether it is a holiday (binary variable feature), the hour of the day, the day of the week, and the month of the year, that is, whether the current moment is a holiday, the time of the day, and the day of the week or month. This information will affect the cooling load applied to the chiller by the object to be cooled, so based on this information, the probability distribution of future cooling loads can be accurately predicted.
[0097] S302, applying a natural gradient boosting algorithm to the meteorological data, the calendar information, and the historical cooling load data to predict the probability distribution of the future cooling load.
[0098] The meteorological data includes the temperature and relative humidity outside the building, and the temperature includes the dry bulb temperature and the wet bulb temperature. The meteorological data collection interval is 30 minutes. The building needs to be cooled by a chiller.
[0099] Applying the natural gradient boosting algorithm NGBoost to the meteorological data, calendar information, and historical cooling load data will predict the probability distribution of the cooling load that the chiller will need to carry in the future. For example, the meteorological data, calendar information, and historical cooling load data are all data at 12 noon. NGBoost predicts the probability distribution of the future cooling load within the hour from 12 noon to 1 pm, where the future cooling load probability is as follows: Figure 4 The normal distribution form shown. Figure 4 The left side of the middle dashed line represents the probability that the future cooling load Q is less than z, which is F(z). The probability distribution of future cooling load is used to characterize the probability of each cooling load that may occur at a certain moment in the future.
[0100] NGBoost can not only output the predicted mean, but also use the distribution (denoted by P θ (y|x)) to express the uncertainty of the prediction. For example, NGBoost can estimate the mean and standard deviation of a normal distribution.
[0101] NGBoost training introduces probabilistic loss functions for optimization, including the log-likelihood function. The log-likelihood function measures the degree of fit between the predicted distribution and the observed data.
[0102]
[0103] Where θ is the parameter of the predicted distribution, including the mean and standard deviation of the normal distribution, and y is the actual cooling load.
[0104] NGBoost uses natural gradient
[0105]
[0106] in, is the distribution P θ The Fisher information carried,
[0107] The NGBoost output of this embodiment is the probability distribution of the cooling load one hour after the current moment, that is, the probability distribution of the cooling load in the future. Among them, u is the mean of the predicted future cooling load, and σ is the standard deviation. The larger it is, the greater the uncertainty.
[0108] S303: Determine, based on the probability distribution, a first predicted probability that the future cooling load is less than the minimum cooling load and a second predicted probability that the future cooling load is less than the maximum cooling load, and use the first predicted probability and the second predicted probability as the predicted probability.
[0109] When the preselected action in step S200 is the closing action, the first predicted probability is recorded as F t+iΔt (Q l ):
[0110] F t+iΔt (Q l )=P(Q t+iΔt <Q l )
[0111] The second prediction probability is denoted as F t+iΔt (Q h ):
[0112] F t+iΔt (Q h )=P(Q t+iΔt <Q h )
[0113] Where t is the current time, the time period to be predicted after the current time is divided into several equal parts, Δt is the duration of each equal part, i is the i-th equal part, Q t+iΔt is the cooling load at the future time t+iΔt, Q t+iΔt <Q h Indicates that the cooling load at the future moment is less than the minimum cooling load Q that the chiller can bear l , indicating that the command supply of the chiller matches the cooling demand of the building.
[0114] When the preselected action in step S200 is the start action, the first predicted probability is recorded as F′ t+iΔt (Q l ):
[0115] F t+iΔt (Q l )=P(Q t+iΔt <Q′ l )
[0116] The second predicted probability is denoted as F′ t+iΔt (Q h ):
[0117] F′ t+iΔt (Q h )=P(Q t+iΔt <Q′ h )
[0118] In one embodiment, when step S200 is a closing action, step S400 includes the following specific steps S401 to S405:
[0119] S401, subtract the first predicted probability from 1 to obtain the first risk value r corresponding to the initial number A1 t+iΔt (A1):
[0120] r t+iΔt (A1)=1-F t+iΔt (Q l )
[0121] S402: The first prediction probability F t+iΔt (Q l ) minus the second predicted probability F t+iΔt (Q h ), and obtain the second risk value r corresponding to the alternative quantity A2 t+iΔt (A2):
[0122] r t+iΔt (A2)=F t+iΔt (Q l )+1-F t+iΔt (Q h )
[0123] When the A1 chiller is turned off, the corresponding first risk value is Figure 5 The shaded part in Figure 5 The horizontal axis is all possible future cooling loads at the future time t+iΔt, and the vertical axis is all possible future cooling loads.
[0124] The probability corresponding to the load.
[0125] When the A2 chiller is turned off, the corresponding second risk value is Figure 6 The shaded part in Figure 6 The horizontal axis is all possible future cooling loads at the future time t+iΔt, and the vertical axis is all possible future cooling loads.
[0126] The probability corresponding to the load.
[0127] S403, accumulating the first risk values generated by using the initial set number A1 at each future moment to obtain a first risk cumulative value r(A1):
[0128] r(A1)=∑ i w i ·r t+iΔt (A1)
[0129] S404, accumulating the second risk values generated by using the candidate quantity A2 at each future moment to obtain a second risk cumulative value r(A2):
[0130] r(A2)=∑ i w i ·r t+iΔt (A2)
[0131] w i is the weight.
[0132] S405, when the first risk cumulative value is less than the second risk cumulative value, the initial quantity A1 is used as the target quantity; when the second risk cumulative value is less than the first risk cumulative value, the alternative quantity is used as the target quantity.
[0133] That is, if r(A1) is less than r(A2), then the initial number A1 is set as the target number, that is, shutting down A1 chillers. In other words, shutting down A1 chillers at the current time t can also ensure that the remaining chillers can meet the cooling needs of the building at the future time t+iΔt.
[0134] If r(A1) is greater than r(A2), then the alternative number A2 is used as the target number, that is, shutting down A2 chillers. In other words, shutting down A2 chillers at the current time t can also ensure that the remaining chillers can meet the cooling demand of the building at the future time t+iΔt.
[0135] In another embodiment, if the preselected action in step S200 is to start the chiller, the risk value r′ corresponding to starting the chillers of the candidate number A2 is calculated. t+iΔt (A2), the risk value r′ corresponding to the initial number A1 of chillers being turned on t+iΔt (A1)
[0136] r t+iΔt (A2) = 1-F t+iΔt (Q′ l )
[0137] r t+iΔt (A1)=F t+iΔt (Q′ l )+1-F t+iΔt (Q′ h )
[0138] Cumulative risk value r′ t+iΔt (A2), we get ∑ i w i (1-F t+iΔt (Q′ l )); accumulate r′ t+iΔt(A1), we get ∑ i w i [F t+iΔt (Q′ l )+1-F t+iΔt (Q′ h )].
[0139] If i w i (1-F t+iΔt (Q′ l )) is less than ∑ i w i [F t+iΔt (Q′ l )+1-F t+iΔt (Q′ h )], then the alternative quantity A2 is taken as the target quantity, that is, A2 chillers are turned on. That is, A2 chillers need to be turned on at the current time t to ensure that all the chillers together can meet the cooling demand of the building at the future time t+iΔt.
[0140] When A2 chillers are turned on, the corresponding risk value r′ t+iΔt (A2) Figure 7 The shaded part in the figure; the risk value r′ corresponding to turning on the chiller A1 t+iΔt (A1) Figure 8 the shadow part.
[0141] If i w i (1-F t+iΔt (Q′ l )) is greater than ∑ i w i [F t+iΔt (Q′ l )+1-F t+iΔt (Q′ h )], then the initial number A1 is taken as the target number, that is, A1 chillers need to be turned on at the current time t to ensure that all the chillers together can meet the cooling demand of the building at the future time t+iΔt.
[0142] In one embodiment, the reliability of the predicted start-up and shutdown of the chiller unit in the present application is determined by the following method:
[0143] The root mean square error (RMSE), mean absolute percentage error (MAPE), and coefficient of variation of the root mean square error (CV-RMSE) were used to evaluate the average error and reliability of the forecast. In addition, the probability integral transform (PIT) was used to test the consistency and stability of the forecast, and the prediction interval average width (PIAW) was used to evaluate the reliability of the forecast.
[0144] In summary, the control method proposed in the present invention can reduce unnecessary switching of chillers. That is, by considering short-term cooling load fluctuations and prediction uncertainties, the control strategy of the present invention can significantly reduce the number of on-off switching of chillers and reduce unnecessary on-off actions in the system.
[0145] The present invention improves the reliability and life of the chiller, that is, the present invention effectively reduces unnecessary on / off actions within 1 hour, which helps to improve the reliability of the chiller and extend its service life.
[0146] The control strategy of the present invention exhibits higher robustness, can effectively cope with the uncertainty of cooling load prediction, and reduce the switching frequency of the chiller when the uncertainty increases.
[0147] The present invention is expected to improve the thermal comfort of end users and ensure that the indoor temperature is stable within a reasonable range through more stable and optimized chiller sequence control.
[0148] This embodiment also provides a chiller control device, such as Fig.10 As shown, the control device includes the following components:
[0149] A quantity construction module 01 is used to determine a preselected action of a chiller and a preset number of the chillers that need to perform the preselected action, wherein the action is used to represent the action of starting and closing the chiller;
[0150] A cooling load calculation module 02, for determining the ultimate cooling load that the chillers can bear after the preset number of chillers perform the preselected actions;
[0151] The probability calculation module 03 is used to obtain the meteorological data at the current moment, and based on the meteorological data, predict the predicted probability that the extreme cooling load matches the future cooling load, where the future cooling load is the cooling load that the chiller needs to bear at a future moment;
[0152] The quantity screening module 04 is used to screen out a target quantity from the preset quantity according to the predicted probability, and control the target quantity of chillers to perform the preset action at the future moment.
[0153] Based on the above embodiments, the present invention further provides a terminal device, whose principle block diagram can be shown as follows: Fig.11 As shown. The terminal device includes a processor, a memory, a network interface, and a display screen connected through a system bus. Among them, the processor of the terminal device is used to provide computing and control capabilities. The memory of the terminal device includes a non-volatile storage medium and an internal memory. The non-volatile storage medium stores an operating system and a computer program. The internal memory provides an environment for the operation of the operating system and the computer program in the non-volatile storage medium. The network interface of the terminal device is used to communicate with an external terminal through a network connection. When the computer program is executed by the processor, a chiller control method is implemented. The display screen of the terminal device can be a liquid crystal display screen or an electronic ink display screen.
[0154] Those skilled in the art will understand that Fig.11 The principle block diagram shown in the figure is only a block diagram of a partial structure related to the scheme of the present invention, and does not constitute a limitation on the terminal device to which the scheme of the present invention is applied. The specific terminal device may include more or fewer components than those shown in the figure, or combine certain components, or have a different arrangement of components.
[0155] In one embodiment, a terminal device is provided, the terminal device comprising a memory, a processor, and a chiller control program stored in the memory and executable on the processor, and when the processor executes the chiller control program, the following operation instructions are implemented:
[0156] Determine a preselected action of the chiller and a preset number of the chillers that need to perform the preselected action, the action being used to represent the action of starting and closing the chiller;
[0157] After the preset number of the chillers perform the preselected actions, determining the ultimate cooling load that the chillers can bear;
[0158] Acquire meteorological data at the current moment, and based on the meteorological data, predict the predicted probability that the extreme cooling load matches the future cooling load, where the future cooling load is the cooling load that the chiller needs to bear at a future moment;
[0159] According to the predicted probability, a target number is screened out from the preset number, and at the future moment, the target number of chillers is controlled to perform the preset action.
[0160] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program, and the computer program can be stored in a non-volatile computer-readable storage medium. When the computer program is executed, it can include the processes of the embodiments of the above-mentioned methods. Among them, any reference to memory, storage, database or other media used in the embodiments provided by the present invention can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration and not limitation, RAM is available in many forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDRSDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM).
[0161] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the embodiments of the present invention.
Claims
1. A chiller control method, It is characterized in that include: Determine a preselected action of the chiller and a preset number of the chillers that need to perform the preselected action, the action being used to represent the action of starting and closing the chiller; After the preset number of the chillers perform the preselected actions, the limit cooling load that the chillers can bear is determined; Acquire meteorological data at the current moment, and based on the meteorological data, predict the predicted probability that the extreme cooling load matches the future cooling load, where the future cooling load is the cooling load that the chiller needs to bear at a future moment; According to the predicted probability, a target number is selected from the preset number, and at the future moment, the target number of chillers is controlled to perform the preselected action; The determining of the preselected action of the chiller and the preset number of the chillers required to perform the preselected action includes: Obtaining the current cooling load carried by the chiller at the current moment, the number of the chillers that have been started and are in the started state at the current moment, and the maximum cooling load that each chiller can carry, and determining a preliminary selection action and an initial number of the chillers that need to perform the preliminary selection action based on the current cooling load, the number of the chillers that have been started and the maximum cooling load; using the primary selection action as the pre-selected action; Subtract the set quantity from the initial quantity to obtain an alternative quantity, and use the initial quantity and the alternative quantity as the preset quantity; The determining of the preliminary action and the initial number of the chillers that need to perform the preliminary action based on the current cooling load, the started number and the maximum cooling load includes: The first intermediate formula is obtained by adding the started number to the set parameter and subtracting 1, and the first intermediate formula is multiplied by the maximum cooling load to construct a first formula; Adding the started quantity to the parameter to obtain a second intermediate formula, and multiplying the second intermediate formula by the maximum cooling load to construct a second formula; The first formula is set to be less than the current cooling load, and the current cooling load is less than or equal to the second formula, so as to construct an inequality; An integer of the parameter satisfying the inequality is determined, and the absolute value of the integer is used as the initial quantity.
2. The chiller control method according to claim 1, It is characterized in that After the preset number of chillers perform the preselected action, determining the limit cooling load that the chillers can bear includes: If the preselected action is a shutdown action, the result of subtracting the initial set number from the started number is multiplied by the maximum cooling load to obtain the minimum cooling load that the chiller can bear; The maximum cooling load that the chiller can bear is obtained by subtracting the alternative number from the started number and multiplying it by the maximum cooling load, and the maximum cooling load and the minimum cooling load are the limit cooling load.
3. The chiller control method according to claim 2, It is characterized in that The predicting, based on the meteorological data, the predicted probability that the extreme cooling load matches the future cooling load comprises: Obtaining calendar information corresponding to the current moment, and obtaining historical cooling load data of the chiller; Applying a natural gradient boosting algorithm to the meteorological data, the calendar information, and the historical cooling load data to predict the probability distribution of the future cooling load; According to the probability distribution, a first predicted probability that the future cooling load is less than the minimum cooling load and a second predicted probability that the future cooling load is less than the maximum cooling load are determined, and the first predicted probability and the second predicted probability are used as the predicted probability.
4. The chiller control method according to claim 3, It is characterized in that The step of selecting a target quantity from the preset quantity according to the predicted probability includes: Subtract the first predicted probability from 1 to obtain a first risk value corresponding to the initial quantity; Subtract the second predicted probability from the result of adding 1 to the first predicted probability to obtain a second risk value corresponding to the number of alternatives; A target quantity is screened out from the initial quantity and the alternative quantities according to the first risk value and the second risk value.
5. The chiller control method according to claim 4, It is characterized in that The step of selecting a target quantity from the initial quantity and the candidate quantities according to the first risk value and the second risk value includes: Accumulating the first risk values generated by using the initial set quantity at each of the future moments to obtain a first risk accumulated value; Accumulating the second risk values generated by using the alternative quantities at each of the future moments to obtain a second risk accumulated value; When the first risk cumulative value is less than the second risk cumulative value, the initial quantity is used as the target quantity; Or, when the second risk cumulative value is less than the first risk cumulative value, the alternative quantity is used as the target quantity.
6. A chiller control device, It is characterized in that The control device comprises the following components: A quantity construction module, used to determine a preselected action of a chiller and a preset number of the chillers that need to perform the preselected action, wherein the action is used to represent the action of starting and closing the chiller; A cooling load calculation module, used for determining the ultimate cooling load that the chillers can bear after the preset number of chillers perform the preselected actions; A probability calculation module, used to obtain meteorological data at the current moment, and based on the meteorological data, predict the predicted probability that the extreme cooling load matches the future cooling load, where the future cooling load is the cooling load that the chiller needs to bear at a future moment; A quantity screening module, used for screening a target quantity from the preset quantity according to the predicted probability, and controlling the target quantity of chillers to perform the preselected action at the future moment; The determining of the preselected action of the chiller and the preset number of the chillers required to perform the preselected action includes: Obtaining the current cooling load carried by the chiller at the current moment, the number of the chillers that have been started and are in the started state at the current moment, and the maximum cooling load that each chiller can carry, and determining a preliminary selection action and an initial number of the chillers that need to perform the preliminary selection action based on the current cooling load, the number of the chillers that have been started and the maximum cooling load; using the primary selection action as the pre-selected action; Subtract the set quantity from the initial quantity to obtain an alternative quantity, and use the initial quantity and the alternative quantity as the preset quantity; The determining of the preliminary action and the initial number of the chillers that need to perform the preliminary action based on the current cooling load, the started number and the maximum cooling load includes: The first intermediate formula is obtained by adding the started number to the set parameter and subtracting 1, and the first intermediate formula is multiplied by the maximum cooling load to construct a first formula; Adding the started quantity to the parameter to obtain a second intermediate formula, and multiplying the second intermediate formula by the maximum cooling load to construct a second formula; The first formula is set to be smaller than the current cooling load, and the current cooling load is smaller than or equal to the second formula, so as to construct an inequality; An integer of the parameter satisfying the inequality is determined, and the absolute value of the integer is used as the initial quantity.
7. A terminal device, It is characterized in that The terminal device includes a memory, a processor, and a chiller control program stored in the memory and executable on the processor. When the processor executes the chiller control program, the steps of the chiller control method according to any one of claims 1 to 5 are implemented.
8. A computer-readable storage medium, It is characterized in that The computer-readable storage medium stores a chiller control program, and when the chiller control program is executed by the processor, the steps of the chiller control method according to any one of claims 1 to 5 are implemented.
