Energy equipment characteristic model construction method and device
Through the construction method of the characteristic model of energy equipment, including curve fitting and error correction, the problem of low modeling efficiency of energy equipment in the existing technology is solved, more accurate and efficient equipment characteristic modeling is achieved, and the production process is optimized.
Patent Information
- Application Number
- CN202410812122.4
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-06-21
- Publication Date
- 2024-11-12
AI Technical Summary
In the energy-saving optimization control scenario in the energy field, it is difficult for the existing technology to effectively model the operating characteristics of energy conversion equipment, resulting in low equipment characteristic modeling efficiency and affecting production efficiency.
By obtaining the sample data when constructing the characteristic model of the target energy equipment, performing curve fitting and error correction, deleting the fitting curve with fitting error, and integrating it with existing sample data to obtain a curve cluster that characterizes the characteristics of the equipment.
It improves the accuracy and efficiency of the characteristic modeling of energy equipment, enhances the understanding of the operating principles of equipment, optimizes the production process, and improves the operating efficiency of enterprises.
Smart Images

Figure CN118940023A_ABST
Abstract
Description
Technical Field
[0001] The present disclosure relates to the technical field of integrated energy systems, and in particular to an energy equipment characteristic model construction method, an energy equipment characteristic model construction device, a storage medium, and a server. Background Art
[0002] In the AI energy-saving optimization control scenario in the energy field, it is usually necessary to model the operating characteristics of various energy conversion equipment, including but not limited to boilers, heat pumps, chillers, etc. In order to improve the operating efficiency of the enterprise, it is sometimes necessary to model the equipment characteristics. Especially for energy equipment, through the equipment model, the enterprise can better understand the operating principles and key characteristics of the equipment, thereby optimizing the production process and improving production efficiency. Summary of the invention
[0003] In view of this, the embodiments of the present disclosure are intended to provide a method for constructing an energy device characteristic model, an apparatus for constructing an energy device characteristic model, a storage medium, and a server.
[0004] The technical solution of the present disclosure is achieved as follows:
[0005] In a first aspect, the present disclosure provides a method for constructing an energy device characteristic model.
[0006] The method for constructing an energy device characteristic model provided by the embodiment of the present disclosure includes:
[0007] Acquire sample data when constructing a characteristic model of a target energy device; wherein the sample data is obtained by combining parameter interval data corresponding to the parameters of the target energy device;
[0008] Based on the sample data when the characteristic model of the target energy device is constructed, curve fitting is performed to obtain a cluster of curves corresponding to the target energy device; wherein each curve corresponds to the device characteristics of the target energy device under one working condition;
[0009] Based on the correlation between the operating condition characteristic value of the target energy device and the parameter, deleting a fitting curve with a fitting error in the curve cluster corresponding to the target energy device;
[0010] Based on the deleted fitting curve with fitting error, combined with existing samples, determine sample data close to the fitting curve with fitting error;
[0011] Based on the sample data close to the fitting curve with fitting error and the sample data corresponding to the undeleted fitting curve in the curve cluster, curve fitting is performed to obtain a curve cluster characterizing the equipment characteristics of the target energy equipment.
[0012] In some embodiments, the step of obtaining sample data when constructing a characteristic model of a target energy device includes:
[0013] Obtain the parameter value range when constructing the characteristic model of the target energy equipment;
[0014] Segmenting the parameter value range by characteristic center to obtain N1 parameter interval data, wherein each parameter interval has a characteristic center;
[0015] The parameter interval data corresponding to all the parameters of the target energy device are combined to obtain sample data when the characteristic model of the target energy device is constructed.
[0016] In some embodiments, the curve fitting is performed based on the sample data when the characteristic model of the target energy device is constructed to obtain a curve cluster corresponding to the target energy device, including:
[0017] Determine the characteristic interval corresponding to each parameter;
[0018] Determining fitting data for curve fitting in each parameter interval based on the characteristic center corresponding to each parameter interval and the characteristic interval corresponding to each parameter;
[0019] Based on the fitting data used for curve fitting in all parameter intervals, curve fitting is performed to obtain a curve cluster corresponding to the target energy equipment.
[0020] In some embodiments, the characteristic intervals corresponding to the parameters include:
[0021] D1_i=Minimum(Maximum((Max_i-Min_i) / CC1_i, CC2_i), CC3_i); wherein D1_i is the characteristic interval corresponding to the i-th parameter; CC1_i is the curve center interval initialization factor corresponding to the i-th parameter; CC2_i is the lower limit of the curve center interval corresponding to the i-th parameter; CC3_i is the upper limit of the curve center interval corresponding to the i-th parameter; CC1_i=(Max_i-Min_i) / N1; CC2_i=CC1_i / 2; CC3_i=2*CC1_i;
[0022] The fitting data used for curve fitting in each parameter interval is (P_1n±D1_i / 2, P_2m±D1_j / 2); wherein (P_1n, P_2m) is the characteristic center corresponding to each parameter interval;
[0023] The step of determining fitting data for curve fitting in each parameter interval based on the characteristic center corresponding to each parameter interval and the characteristic interval corresponding to each parameter includes:
[0024] Based on the characteristic center (P_1n, P_2m) corresponding to each parameter interval and the characteristic interval D1_i corresponding to each parameter, the fitting data (P_1n±D1_i / 2, P_2m±D1_j / 2) used for curve fitting in each parameter interval is determined.
[0025] In some embodiments, based on the correlation between the operating condition characteristic value of the target energy device and the parameter, deleting the fitting curve with fitting error in the curve cluster corresponding to the target energy device includes:
[0026] Determining the theoretical intersection of curves between the fitting curves based on the correlation between the operating characteristic value of the target energy equipment and the parameters within the parameter intervals;
[0027] Based on the theoretical intersection of the fitting curves and the actual intersection of the fitting curves, the fitting curves with fitting errors are deleted from the curve cluster corresponding to the target energy equipment.
[0028] In some embodiments, the step of determining sample data close to the fitting curve with fitting errors based on the deleted fitting curve with fitting errors and combining existing samples includes:
[0029] Determine two sample curves that are closest to the characteristic center of the fitting curve with fitting error among the existing samples;
[0030] Based on the operating condition characteristic value of the target energy equipment corresponding to the fitting curve with fitting error, parameter sampling is performed on the two sample curves to obtain parameter sample data of the operating condition characteristic value corresponding to the fitting curve with fitting error;
[0031] Performing curve fitting based on parameter sample data of operating condition characteristic values corresponding to the fitting curve with fitting error to obtain a sample fitting curve;
[0032] The operating condition characteristic value of the target energy equipment corresponding to the fitting curve with the fitting error is substituted into the sample fitting curve to obtain sample data close to the fitting curve with the fitting error.
[0033] In some embodiments, based on the theoretical intersection of the curves between the fitting curves and the actual intersection of the fitting curves, deleting the fitting curve with the fitting error in the curve cluster corresponding to the target energy device includes:
[0034] Among all the curve intersections, the curve that crosses the same curve the maximum number of times is determined as the target curve;
[0035] The target curve is determined to be a fitting curve with a fitting error, and is deleted.
[0036] In a second aspect, the present disclosure provides an energy device characteristic model construction device, comprising:
[0037] A data acquisition module, used to acquire sample data when constructing a characteristic model of a target energy device; wherein the sample data is obtained by combining parameter interval data corresponding to the parameters of the target energy device;
[0038] A first curve fitting module is used to perform curve fitting based on sample data when the characteristic model of the target energy device is constructed, so as to obtain a curve cluster corresponding to the target energy device; wherein each curve corresponds to a device characteristic of the target energy device under one working condition;
[0039] A curve deletion module, used for deleting a fitting curve with a fitting error in a curve cluster corresponding to the target energy device based on the correlation between the operating condition characteristic value of the target energy device and the parameter;
[0040] A sample data generating module, used to determine sample data close to the fitting curve with fitting error based on the deleted fitting curve with fitting error and in combination with existing samples;
[0041] The second curve fitting module is used to perform curve fitting based on the sample data close to the fitting curve with fitting error and the sample data corresponding to the undeleted fitting curve in the curve cluster to obtain a curve cluster characterizing the equipment characteristics of the target energy equipment.
[0042] In a third aspect, the present disclosure provides a computer-readable storage medium on which a control program for an air compressor system is stored. When the control program for the air compressor system is executed by a processor, the energy equipment characteristic model construction method described in the first aspect is implemented.
[0043] In a fourth aspect, the present disclosure provides a server comprising a memory, a processor, and a control program for an air compressor system stored in the memory and executable on the processor. When the processor executes the control program for the air compressor system, the energy equipment characteristic model construction method described in the first aspect is implemented.
[0044] According to the method for constructing an energy device characteristic model of an embodiment of the present disclosure, the method includes: obtaining sample data when constructing the characteristic model of the target energy device; wherein the sample data is obtained by combining parameter interval data corresponding to the parameters of the target energy device; based on the sample data when constructing the characteristic model of the target energy device, performing curve fitting to obtain a curve cluster corresponding to the target energy device; wherein each curve corresponds to the device characteristic of the target energy device under a working condition; based on the correlation between the working condition characteristic value and the parameter of the target energy device, deleting the fitting curve with fitting error in the curve cluster corresponding to the target energy device; based on the deleted fitting curve with fitting error, combined with existing samples, determining sample data similar to the fitting curve with fitting error; based on the sample data similar to the fitting curve with fitting error and the sample data corresponding to the undeleted fitting curve in the curve cluster, performing curve fitting to obtain a curve cluster characterizing the device characteristics of the target energy device. In this application, curve fitting is performed on sample data corresponding to the parameters of the target energy device to obtain a curve cluster corresponding to the target energy device, and then the curve cluster is corrected, and sample data close to the fitting curve with fitting error is determined based on the deleted fitting curve and combined with the existing samples, and sample integration is performed to improve the validity of the sample data. Then, curve fitting is performed on the integrated sample data to obtain a curve cluster that can characterize the device characteristics of the target energy device.
[0045] Additional aspects and advantages of the present disclosure will be given in part in the following description and in part will be obvious from the following description or learned through practice of the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS
[0046] Figure 1 is a flow chart of a method for constructing an energy device characteristic model according to an exemplary embodiment;
[0047] Figure 2 is a flow chart of curve fitting in the process of building an energy device characteristic model according to an exemplary embodiment;
[0048] Figure 3 is a flow chart of initializing the center of a curve in a process of constructing an energy device characteristic model according to an exemplary embodiment;
[0049] Figure 4 is a flow chart of a preliminary fitting curve in a process of building an energy device characteristic model according to an exemplary embodiment;
[0050] Figure 5 is a flow chart of a correction fitting curve in a process of building an energy device characteristic model according to an exemplary embodiment;
[0051] Figure 6is a schematic diagram of a fitted curve cluster after correction of an energy device characteristic model construction process according to an exemplary embodiment;
[0052] Figure 7 is a synthetic curve flow chart in a process of constructing an energy device characteristic model according to an exemplary embodiment;
[0053] Figure 8 is a flow chart of an example of initializing a curve center in a process of building an energy device characteristic model according to an exemplary embodiment;
[0054] Fig. 9 is a flowchart of a preliminary fitting curve example in a process of building an energy device characteristic model according to an exemplary embodiment;
[0055] Fig.10 is a schematic diagram of a cluster of preliminary fitting example curves in a process of building an energy device characteristic model according to an exemplary embodiment;
[0056] Fig.11 is a flow chart of an example of correcting a fitting curve in a process of building an energy device characteristic model according to an exemplary embodiment;
[0057] Fig.12 is a schematic diagram of a cluster of correction fitting example curves in a process of building an energy device characteristic model according to an exemplary embodiment;
[0058] Fig.13 is a flow chart of a synthetic example curve in a process of building an energy device characteristic model according to an exemplary embodiment;
[0059] Fig.14 It is a schematic diagram of the structure of an energy device characteristic model building device according to an exemplary embodiment. DETAILED DESCRIPTION
[0060] Embodiments of the present disclosure are described in detail below, examples of which are shown in the accompanying drawings, wherein the same or similar reference numerals throughout represent the same or similar elements or elements having the same or similar functions. The embodiments described below with reference to the accompanying drawings are exemplary and are intended to be used to explain the present disclosure, and should not be construed as limiting the present disclosure.
[0061] In the AI energy-saving optimization control scenario in the energy field, it is usually necessary to model the operating characteristics of various energy conversion equipment, including but not limited to boilers, heat pumps, chillers, etc. In order to improve the operating efficiency of the enterprise, it is sometimes necessary to model the equipment characteristics. Especially for energy equipment, through the equipment model, the enterprise can better understand the operating principles and key characteristics of the equipment, thereby optimizing the production process and improving production efficiency.
[0062] In view of the above situation, the present disclosure provides a method for constructing an energy device characteristic model. Figure 1 FIG. 1 is a flow chart of a method for constructing an energy device characteristic model according to an exemplary embodiment. Figure 1 As shown, the energy equipment characteristic model construction method includes:
[0063] Step 10, obtaining sample data when constructing the characteristic model of the target energy device; wherein the sample data is obtained by combining parameter interval data corresponding to the parameters of the target energy device;
[0064] Step 11: Perform curve fitting based on sample data when constructing the characteristic model of the target energy device to obtain a cluster of curves corresponding to the target energy device; wherein each curve corresponds to the device characteristics of the target energy device under one working condition;
[0065] Step 12: based on the correlation between the operating condition characteristic value of the target energy device and the parameter, deleting the fitting curve with fitting error in the curve cluster corresponding to the target energy device;
[0066] Step 13: Based on the deleted fitting curve with fitting error and in combination with the existing samples, determine sample data close to the fitting curve with fitting error;
[0067] Step 14: Perform curve fitting based on the sample data close to the fitting curve with fitting error and the sample data corresponding to the undeleted fitting curve in the curve cluster to obtain a curve cluster representing the device characteristics of the target energy device.
[0068] In this exemplary embodiment, the data characterizing the working condition of the target energy device includes working condition characteristic values and parameters corresponding to the working condition characteristic values. The working condition characteristic values and parameters corresponding to the working condition characteristic values are used to characterize various working conditions of the target energy device.
[0069] In this exemplary embodiment, the correlation between the operating characteristic value of the target energy device and the parameter includes a positive correlation or a negative correlation. If the correlation between the operating characteristic value of the target energy device and the parameter within the parameter interval is positively correlated and the positive correlation value is 1, or if the correlation between the operating characteristic value of the target energy device and the parameter is negatively correlated and the negative correlation value is -1, then theoretically the two characteristic curves will not intersect, and at this time, the curve generated in the previous step can be cross-checked to eliminate abnormal curves.
[0070] In this exemplary embodiment, the target energy equipment may include a chiller, an air source heat pump, a water source heat pump, a hot water boiler, a steam boiler, and the like. When the target energy equipment is a chiller, the operating condition characteristic values of the target energy equipment may include COP (Coefficient of Performance) and refrigeration load, and the parameters of the target energy equipment include chilled water flow, cooling water flow, chilled water inlet temperature, and cooling water inlet temperature; when the target energy equipment is an air source heat pump, the operating condition characteristic values of the target energy equipment may include COP, refrigeration load / heating load, and the parameters of the target energy equipment include circulating water flow, inlet water temperature, and ambient temperature; when the target energy equipment is a water-ground source heat pump, the operating condition characteristic values of the target energy equipment may include COP, refrigeration load / heating load, and the parameters of the target energy equipment include user-side water flow, source-side water flow, user-side inlet water temperature, and source-side inlet water temperature; when the target energy equipment is a hot water boiler, the operating condition characteristic values of the target energy equipment may include thermal efficiency and heating load, and the parameters of the target energy equipment include circulating water flow and inlet water temperature; when the target energy equipment is a steam boiler, the operating condition characteristic values of the target energy equipment may include steam unit consumption and steam load, and the parameters of the target energy equipment include outlet steam pressure. Among them, the operating condition characteristic values corresponding to different target energy equipment may be different, and the parameters corresponding to different target energy equipment may also be different.
[0071] According to the method for constructing an energy device characteristic model of an embodiment of the present disclosure, the method includes: obtaining sample data when constructing the characteristic model of the target energy device; wherein the sample data is obtained by combining parameter interval data corresponding to the parameters of the target energy device; based on the sample data when constructing the characteristic model of the target energy device, performing curve fitting to obtain a curve cluster corresponding to the target energy device; wherein each curve corresponds to the device characteristic of the target energy device under a working condition; based on the correlation between the working condition characteristic value and the parameter of the target energy device, deleting the fitting curve with fitting error in the curve cluster corresponding to the target energy device; based on the deleted fitting curve with fitting error, combined with existing samples, determining sample data similar to the fitting curve with fitting error; based on the sample data similar to the fitting curve with fitting error and the sample data corresponding to the undeleted fitting curve in the curve cluster, performing curve fitting to obtain a curve cluster characterizing the device characteristics of the target energy device. In this application, curve fitting is performed on sample data corresponding to the parameters of the target energy device to obtain a curve cluster corresponding to the target energy device, and then the curve cluster is corrected, and sample data close to the fitting curve with fitting error is determined based on the deleted fitting curve and combined with the existing samples, and sample integration is performed to improve the validity of the sample data. Then, curve fitting is performed on the integrated sample data to obtain a curve cluster that can characterize the device characteristics of the target energy device.
[0072] Figure 2 FIG. 1 is a flow chart of curve fitting in the process of constructing an energy device characteristic model according to an exemplary embodiment. Figure 2 As shown in Figure 1, the curve fitting process in the process of building the energy equipment characteristic model includes:
[0073] Step 20, initializing the center of the curve;
[0074] Step 21, preliminary fitting curve;
[0075] Step 22, correct the fitting curve;
[0076] Step 23, synthesizing the fitting curve;
[0077] Step 24: Predict the curve results.
[0078] In some embodiments, the step of obtaining sample data when constructing a characteristic model of a target energy device includes:
[0079] Obtain the parameter value range when constructing the characteristic model of the target energy equipment;
[0080] Segmenting the parameter value range by characteristic center to obtain N1 parameter interval data, wherein each parameter interval has a characteristic center;
[0081] The parameter interval data corresponding to all the parameters of the target energy device are combined to obtain sample data when the characteristic model of the target energy device is constructed.
[0082] In this exemplary embodiment, the sample data is obtained by combining the parameter interval data corresponding to the parameters of the target energy device. When obtaining the sample data when the characteristic model of the target energy device is constructed, the parameter value range when the characteristic model of the target energy device is constructed can be obtained; the parameter value range is segmented by the characteristic center to obtain N1 parameter interval data; each parameter interval has a characteristic center; the parameter interval data corresponding to all the parameters of the target energy device are combined to obtain the sample data when the characteristic model of the target energy device is constructed. Figure 3 FIG. 1 is a flow chart of initializing the center of a curve in the process of constructing an energy device characteristic model according to an exemplary embodiment. Figure 3 As shown in the figure, the curve center initialization process in the process of building the energy equipment characteristic model includes:
[0083] Step 30, start;
[0084] Step 31, calculate the maximum value Max_i and the minimum value Min_i (i=1, 2, 3...N) of each dimension feature;
[0085] Step 32, according to the number of feature centers in each dimension N1, the data segment from the minimum value Min_i to the maximum value Max_i is divided into N1 segments, and the endpoints P_ij of each segment are obtained; where (i=1, 2, 3...N, j=1, 2, 3...N1);
[0086] Step 33, according to the data endpoints P_ij of each feature, the features are combined with each other to obtain the center vector of the ∏N1 curve, referred to as C_k (k=1, 2, 3...∏N1);
[0087] Step 34, end.
[0088] In this exemplary embodiment, for example, the maximum value and minimum value corresponding to each dimension of parameter data are obtained; according to the number of characteristic centers N1 of each dimension of parameter data, each dimension of parameter data is equally divided into data segments to obtain each parameter interval data including the endpoints of each segment of data; according to each parameter interval data of each dimension of data, the characteristics are combined with each other in units to obtain a parameter interval data combination including N1 to the power of N1; wherein, a parameter interval data combination is a curve center vector; and a curve center vector corresponds to a fitting curve. The parameter interval data combination of N1 to the power of N1 is the sample data for constructing the characteristic model of the target energy device.
[0089] The above samples are sample data based on two-dimensional parameter data. The number of parameter interval data combinations in the sample data is N1 to the power of N1; when constructing samples based on three-dimensional parameter data, if each parameter interval is N1 data interval, the number of parameter interval data combinations contained in the constructed sample data is N1 to the power of N1, and then N1 again.
[0090] In some embodiments, the curve fitting is performed based on the sample data when the characteristic model of the target energy device is constructed to obtain a curve cluster corresponding to the target energy device, including:
[0091] Determine the characteristic interval corresponding to each parameter;
[0092] Determining fitting data for curve fitting in each parameter interval based on the characteristic center corresponding to each parameter interval and the characteristic interval corresponding to each parameter;
[0093] Based on the fitting data used for curve fitting in all parameter intervals, curve fitting is performed to obtain a curve cluster corresponding to the target energy equipment.
[0094] In this exemplary embodiment, when performing curve fitting through various parameter interval data combinations in the sample data, data screening can be performed on various parameter interval data combinations first to screen out interval valid data, and then curve fitting can be performed based on the interval valid data.
[0095] In this exemplary embodiment, Figure 4 FIG. 1 is a flow chart of a preliminary fitting curve in the process of building an energy device characteristic model according to an exemplary embodiment. Figure 4 As shown in the figure, the preliminary curve fitting process in the process of building the energy equipment characteristic model includes:
[0096] Step 40, start;
[0097] Step 41, calculate the maximum value Max_i and the minimum value Min_i (i=1, 2, 3...N) of each dimension feature;
[0098] Step 42, calculating the feature interval D1_i of each dimension;
[0099] Step 43: According to the curve center vector C_ki and the characteristic interval D1_i, sample data within the range of C_ki±D1_i / 2 is selected. At this time, the data only includes the load rate and the target characteristic value. The formula y=ax 2 +bx+c are fitted to obtain the undetermined parameters a_k, b_k, c_k (k=1, 2, 3...∏N1_i) corresponding to the center of each curve; among which, the load rate and the target characteristic value are both characteristic values of the working condition;
[0100] Step 44, end.
[0101] Among them, the characteristic intervals corresponding to the parameters include:
[0102] D1_i=Minimum(Maximum((Max_i-Min_i) / CC1_i, CC2_i), CC3_i); wherein D1_i is the characteristic interval corresponding to the i-th parameter; CC1_i is the curve center interval initialization factor corresponding to the i-th parameter; CC2_i is the lower limit of the curve center interval corresponding to the i-th parameter; CC3_i is the upper limit of the curve center interval corresponding to the i-th parameter; CC1_i=(Max_i-Min_i) / N1; CC2_i=CC1_i / 2; CC3_i=2*CC1_i;
[0103] The fitting data used for curve fitting in each parameter interval is (P_1n±D1_i / 2, P_2m±D1_j / 2); wherein (P_1n, P_2m) is the characteristic center corresponding to each parameter interval;
[0104] The step of determining fitting data for curve fitting in each parameter interval based on the characteristic center corresponding to each parameter interval and the characteristic interval corresponding to each parameter includes:
[0105] Based on the characteristic center (P_1n, P_2m) corresponding to each parameter interval and the characteristic interval D1_i corresponding to each parameter, the fitting data (P_1n±D1_i / 2, P_2m±D1_j / 2) used for curve fitting in each parameter interval is determined.
[0106] In this exemplary embodiment, Maximum9() is the maximum value between the two, and Minimum() is the minimum value between the two. After obtaining the interval valid data in the parameter interval data combination corresponding to each parameter, curve fitting can be performed based on the interval valid data in the parameter interval data combination corresponding to each parameter to obtain the curve cluster corresponding to the target energy device. If it is a curve fitting corresponding to a two-dimensional parameter, a two-dimensional curve fitting can be performed to obtain each fitting curve y=ax 2 +bx+c corresponding to the unknown parameters a, b, c, thus obtaining the fitting curve.
[0107] In some embodiments, based on the correlation between the operating condition characteristic value of the target energy device and the parameter, deleting the fitting curve with fitting error in the curve cluster corresponding to the target energy device includes:
[0108] Determining the theoretical intersection of curves between the fitting curves based on the correlation between the operating characteristic value of the target energy equipment and the parameters within the parameter intervals;
[0109] Based on the theoretical intersection of the fitting curves and the actual intersection of the fitting curves, the fitting curves with fitting errors are deleted from the curve cluster corresponding to the target energy equipment.
[0110] In this exemplary embodiment, based on the correlation between the operating condition characteristic value of the target energy equipment and the parameters in each parameter interval, determining the curve theoretical intersection between each fitting curve includes:
[0111] If the correlation between the operating characteristic value of the target energy equipment and the parameters in the two parameter intervals is positively correlated, and the positive correlation value is 1, or the correlation between the operating characteristic value of the target energy equipment and the parameters in each parameter interval is negatively correlated, and the negative correlation value is -1, then it can be determined that the fitting curves corresponding to the two parameter intervals are theoretically non-intersecting, otherwise the fitting curves corresponding to the two parameter intervals may intersect in theory. When the two fitting curves are theoretically non-intersecting, but the fitting curves obtained in the actual fitting process intersect, it means that at least one of the two fitting curves has a fitting error.
[0112] The fitting curve with fitting error can be determined by the intersection of a fitting curve and the other fitting curves, and then the fitting curve with fitting error is deleted.
[0113] The operating characteristic values of the target energy equipment include load rate and COP performance coefficient;
[0114] The determining of the curve theoretical intersection between each fitting curve based on the correlation between the operating condition characteristic value of the target energy equipment and the parameters within each parameter interval includes:
[0115] Based on the correlation between the COP performance coefficient and the parameters at the same load rate within each parameter interval, the curve theoretical intersection between each fitting curve is determined.
[0116] The determining of the theoretical intersection of the curves between the fitting curves based on the correlation between the COP performance coefficient and the parameters at the same load rate within each parameter interval includes:
[0117] Determine the curve center vector (P_1A, P_2A) of curve A, the curve center vector (P_1B, P_2B) of curve B, the chilled water flow characteristic correlation = DD3_1, and the cooling water flow characteristic correlation = DD3_2; wherein the chilled water flow characteristic correlation characterizes the correlation between the COP performance coefficient and the chilled water flow; the cooling water flow characteristic correlation characterizes the correlation between the COP performance coefficient and the cooling water flow; P_1A and P_2A are both characteristic centers of the parameters of curve A, and P_1B and P_2B are characteristic centers of the parameters of curve B; for example, P_1A is the characteristic center of the parameter chilled water flow of curve A, P_2A is the characteristic center of the parameter cooling water flow of curve A, P_1B is the characteristic center of the parameter chilled water flow of curve B, and P_2B is the characteristic center of the parameter cooling water flow of curve B;
[0118] Determine F1 = (P_1A-P_1B)*DD3_1, F2 = (P_2A-P_2B)*DD3_2;
[0119] Determine whether F1 is equal to F2;
[0120] If F1 is equal to F2, it means that the curves are consistent and theoretically there is no intersection between the curves;
[0121] If F1 is not equal to F2, it means that the curves are inconsistent and the curves can theoretically intersect.
[0122] Figure 5 FIG. 1 is a flow chart of a correction fitting curve in a process of constructing an energy device characteristic model according to an exemplary embodiment. Figure 5 As shown in the figure, the process of correcting the fitting curve in the process of building the energy equipment characteristic model includes:
[0123] Step 50, start;
[0124] Step 51, counting the number of curves n1;
[0125] Step 52, traverse each curve, count the number of curves among the remaining curves that intersect the target curve in the load rate [DD1, DD2] interval and are consistent with the target curve, and obtain n_i (i=1, 2, 3...N);
[0126] Step 53: determine whether max(n_i) is greater than 0;
[0127] Step 54, if max(n_i)>0, delete the curve with the largest number of intersections, and then loop through steps 51 to 54;
[0128] Step 55: If max(n_i) is not greater than 0, end.
[0129] Figure 6 FIG. 1 is a schematic diagram of a fitted curve cluster after correction of the energy device characteristic model construction process according to an exemplary embodiment. Figure 6 As shown, after curve correction, we can get Figure 6 The curve cluster shown is after deleting the fitting curve with fitting error. Figure 6 The horizontal axis is the load factor and the vertical axis is the COP.
[0130] In some embodiments, the step of determining sample data close to the fitting curve with fitting errors based on the deleted fitting curve with fitting errors and combining existing samples includes:
[0131] Determine two sample curves that are closest to the characteristic center of the fitting curve with fitting error among the existing samples;
[0132] Based on the operating condition characteristic value of the target energy equipment corresponding to the fitting curve with fitting error, parameter sampling is performed on the two sample curves to obtain parameter sample data of the operating condition characteristic value corresponding to the fitting curve with fitting error;
[0133] Performing curve fitting based on parameter sample data of operating condition characteristic values corresponding to the fitting curve with fitting error to obtain a sample fitting curve;
[0134] The operating condition characteristic value of the target energy equipment corresponding to the fitting curve with the fitting error is substituted into the sample fitting curve to obtain sample data close to the fitting curve with the fitting error.
[0135] In this exemplary embodiment, since the sample data corresponding to the fitting curve with fitting error is also part of the overall sample data, the sample data corresponding to the deleted fitting curve can be reused. For example, in the existing samples, the two sample curves closest to the feature center of the fitting curve with fitting error are determined;
[0136] Based on the operating condition characteristic value of the target energy equipment corresponding to the fitting curve with fitting error, parameter sampling is performed on the two sample curves to obtain parameter sample data of the operating condition characteristic value corresponding to the fitting curve with fitting error;
[0137] Performing curve fitting based on parameter sample data of operating condition characteristic values corresponding to the fitting curve with fitting error to obtain a sample fitting curve;
[0138] The operating condition characteristic value of the target energy equipment corresponding to the fitting curve with the fitting error is substituted into the sample fitting curve to obtain sample data close to the fitting curve with the fitting error.
[0139] In this exemplary embodiment, Figure 7 FIG. 1 is a flow chart of a synthetic curve in a process of constructing an energy device characteristic model according to an exemplary embodiment. Figure 7 As shown in the figure, the synthesis curve process in the process of building the energy equipment characteristic model includes:
[0140] Step 70, start;
[0141] Step 71, counting the curve samples deleted due to correction, there are M curves in total, each curve has n_k samples, (k=1, 2, 3...N);
[0142] Step 72, let i=1;
[0143] Step 73: Based on the characteristic center of the i-th curve, find two curves with the smallest Euclidean distance from its center, which are curve A and curve B;
[0144] Step 74: randomly collect EE1 load rates x_j (j=1, 2, 3...EE1) simultaneously in the load rate interval [DD1, DD2];
[0145] Step 75: According to x_j, collect corresponding target values y_A_j and y_B_j from curve A and curve B respectively;
[0146] Step 76: According to the N-dimensional eigenvalue, y_A_j is used as the target value, and the center of curve A is used as the N-dimensional independent variable to construct a sample. Similarly, a sample is constructed for curve B. According to the two samples, a linear equation of N variables is used for fitting to obtain fitting coefficients F_i (i=1, 2, 3...N);
[0147] Step 77, according to the fitting coefficient, calculate the target value when the independent variable is the characteristic center of the i-th curve, obtain the sample (x_j, y_j), and merge it with the original sample of the curve;
[0148] Step 78: Use the formula y=ax 2 +bx+c to fit and obtain the unknown parameters a_i, b_i, c_i corresponding to the center of each curve;
[0149] Step 79, determine whether i is greater than M;
[0150] Step 80: If i is not greater than M, then i=i+1, and steps 72 to 79 are executed in a loop;
[0151] Step 81: If i is greater than M, then end.
[0152] Specifically, after the correction of the fitting curve is completed, some fitting failure samples are obtained. For example, in the range of the center of the curve where the chilled water flow rate is 100 and the cooling water flow rate is 150, there are some fitting failure samples. Now it is necessary to use the fitted curve to use these samples.
[0153] First, among the generated curves, find the curve whose center is closest to (100,150), for example, find curve A whose center is (110,160) and curve B whose center is (90,140).
[0154] Secondly, we need to sample curves A and B by traversing the load rate, and then perform linear regression to generate samples. For example, we need to sample samples with a load rate of 40. We get sample COP=6 from curve A and sample COP=5 from curve B. According to the center of the corresponding curve, we assemble A: COP=6, chilled water flow=110, cooling water flow=160; B: COP=5, chilled water flow=90, cooling water flow=140.
[0155] Then, it is necessary to use COP as the target value and the center of the curve as the independent variable for linear regression fitting. The fitting formula is: y=k1·x1+k2·x2. The above two samples are used to fit the formula, and the parameters k1 and k2 to be determined are determined. Then, the center of the target curve x1=100, x2=150 is substituted, and COP=4 can be obtained. At this time, COP=4 and load rate=40 become the sample with the center of the curve (100,150). In this way, samples under different load rates can be obtained. Finally, the characteristic curve under the center of the curve can be fitted based on these samples.
[0156] In some embodiments, based on the theoretical intersection of the curves between the fitting curves and the actual intersection of the fitting curves, deleting the fitting curve with the fitting error in the curve cluster corresponding to the target energy device includes:
[0157] Among all the curve intersections, the curve that crosses the same curve the maximum number of times is determined as the target curve;
[0158] The target curve is determined to be a fitting curve with a fitting error, and is deleted.
[0159] In this exemplary embodiment, when determining a fitting curve with a fitting error, it is possible to determine the fitting curves among all the fitting curves that should not intersect with the same fitting curve, that is, the fitting curves with a positive correlation value of 1 or a negative correlation value of -1; then, the number of curves that actually intersect among the fitting curves with a positive correlation value of 1 or a negative correlation value of -1 is determined; among the number of curves that should not intersect but do intersect for each fitting curve, the fitting curve with the largest number of intersecting curves is the fitting curve with a fitting error; the fitting curve with a fitting error can be deleted.
[0160] Figure 8 FIG. 1 is a flow chart showing an example of initializing the center of a curve in the process of constructing an energy device characteristic model according to an exemplary embodiment. Figure 8 As shown in FIG. 1 , the curve center initialization example process in the process of building the energy equipment characteristic model includes:
[0161] Step 82, start;
[0162] Step 83, calculating the maximum value Max_1 and the minimum value Min_1 of the chilled water flow, and calculating the maximum value Max_2 and the minimum value Min_2 of the cooling water flow;
[0163] Step 84, according to the number of characteristic centers N1, Max_1 and Min_1, the characteristic-chilled water flow is evenly divided to obtain P_1j (j=1, 2, 3...N1); according to the number of characteristic centers N1, Max_2 and Min_2, the characteristic-chilled water flow is evenly divided to obtain P_2j (j=1, 2, 3...N1);
[0164] Step 85: Combine the data endpoints P_1j and P_2j to obtain N1*N1 curve center vectors (P_1n, P_2m), (n=1, 2, 3...N1) (m=1, 2, 3...N1).
[0165] Step 86, end.
[0166] Fig. 9 FIG. 1 is a flowchart of an example of a preliminary fitting curve in the process of building an energy device characteristic model according to an exemplary embodiment. Fig. 9 As shown in the figure, the initial curve fitting example process in the process of building the energy equipment characteristic model includes:
[0167] Step 90, start;
[0168] Step 91, calculating the maximum value Max_1 and the minimum value Min_1 of the chilled water flow, and calculating the maximum value Max_2 and the minimum value Min_2 of the cooling water flow;
[0169] Step 92: Calculate the chilled water flow characteristic interval
[0170] D1_1=Minimum(Maximum((Max_i-Min_i) / CC1_i, CC2_i), CC3_i);
[0171] Calculate cooling water flow characteristic interval
[0172] D1_2=Minimum(Maximum((Max_i-Min_i) / CC1_i, CC2_i), CC3_i);
[0173] Step 93: According to the curve center vector (P_1n, P_2m), and the characteristic intervals DD1 and DD2, select the samples within the range of (P_1n±D1_i / 2, P_2m±D1_j / 2), take the load rate as the independent variable and COP as the dependent variable, and use the formula y=ax 2 +bx+c to fit and obtain the unknown parameters a_k, b_k, c_k (k=1, 2, 3...N1*N1) corresponding to the center of each curve;
[0174] Step 94, end.
[0175] Fig.10FIG. 1 is a schematic diagram of a cluster of preliminary fitting example curves in the process of building an energy device characteristic model according to an exemplary embodiment. Fig.10 As shown, there are intersections between many fitting curves in the preliminary fitting example curve cluster. When there are intersections between fitting curves that should not be crossed, the corresponding cross-fitting curves may have fitting errors.
[0176] Fig.11 FIG. 1 is a flow chart showing an example of a correction fitting curve in a process of constructing an energy device characteristic model according to an exemplary embodiment. Fig.11 As shown in FIG. 1 , the example process of correcting the fitting curve in the process of building the energy equipment characteristic model includes:
[0177] Step 100, consistency determination between curves A and B begins;
[0178] Step 101, determine the curve center vector (P_1A, P_2A) of curve A, the curve center vector (P_1B, P_2B) of curve B, the chilled water flow characteristic correlation = DD3_1, and the cooling water flow characteristic correlation = DD3_2;
[0179] Step 102, determine F1 = (P_1A-P_1B)*DD3_1, F2 = (P_2A-P_2B)*DD3_2;
[0180] Step 103, determine whether F1 is equal to F2;
[0181] Step 104: If F1 is equal to F2, it means that the curves are consistent;
[0182] Step 105: If F1 is not equal to F2, it means that the curves are inconsistent;
[0183] Step 106, the consistency determination between curves A and B is completed;
[0184] Step 107, traverse each curve, count the number of curves among the remaining curves that intersect the target curve in the load rate [DD1, DD2] interval and are consistent with the target curve, and obtain n_i (i=1, 2, 3...N);
[0185] Step 108: Determine whether max(n_i) is greater than 0;
[0186] Step 109, if max(n_i)>0, delete the curve with the largest number of intersections, and then execute steps 107 to 109 in a loop;
[0187] Step 110: If max(n_i) is not greater than 0, end.
[0188] Fig.12FIG. 1 is a schematic diagram of a cluster of curves of a modified fitting example in the process of constructing an energy device characteristic model according to an exemplary embodiment. Fig.12 As shown, the modified fitting example curve cluster is a curve cluster obtained by deleting the curves with fitting errors from the preliminary fitting example curve cluster.
[0189] Fig.13 FIG. 1 is a flow chart of a synthetic example curve in a process of constructing an energy device characteristic model according to an exemplary embodiment. Fig.13 As shown in the figure, the synthetic instance curve process in the process of building the energy equipment characteristic model includes:
[0190] Step 130, start;
[0191] Step 131, counting the curve samples deleted due to correction, there are M curves in total, each curve has n_k samples, (k=1, 2, 3...N);
[0192] Step 132, let i=1;
[0193] Step 133: based on the center of the i-th curve (chilled water flow, cooling water flow), find two curves with the smallest normalized Euclidean distance from the center, which are curve A and curve B;
[0194] Step 134, randomly collect EE1 load rates x_j (j=1, 2, 3...EE1) simultaneously in the load rate interval [DD1, DD2];
[0195] Step 135: According to x_j, collect corresponding COP values y_A_j and y_B_j from curve A and curve B respectively;
[0196] Step 136, construct a sample according to the center of curve A (chilled water flow A, cooling water flow A) and the COP value y_A_j under the load rate x_j; construct a sample according to the center of curve B (chilled water flow B, cooling water flow B) and the COP value y_B_j under the load rate x_j; for these two samples, use the curve center as the independent variable and COP as the dependent variable to fit the formula y=a1·x1+a2·x2, determine the parameters a1 and a2, and bring the center of the i-th curve (chilled water flow, cooling water flow) into the formula to obtain the COP value of the curve under the load rate x_j, so that samples (x_j, COP_j) under different load rates are obtained, (j=1, 2, 3...EE1);
[0197] Step 137: merge the samples (x_j, COP_j) at different load rates with the original samples of the curve, using the formula y=ax 2+bx+c to fit and obtain the unknown parameters a_i, b_i, c_i corresponding to the center of each curve;
[0198] Step 138, determine whether i is greater than M;
[0199] Step 139: If i is not greater than M, set i=i+1 and execute steps 132 to 139 in a loop;
[0200] Step 140: If i is greater than M, then end.
[0201] The present disclosure provides an energy equipment characteristic model construction device, comprising:
[0202] The data acquisition module 30 is used to acquire sample data when constructing the characteristic model of the target energy device; wherein the sample data is obtained by combining parameter interval data corresponding to the parameters of the target energy device;
[0203] A first curve fitting module 31 is used to perform curve fitting based on sample data when the characteristic model of the target energy device is constructed, so as to obtain a cluster of curves corresponding to the target energy device; wherein each curve corresponds to a device characteristic of the target energy device under one working condition;
[0204] A curve deletion module 32, configured to delete a fitting curve having a fitting error in a curve cluster corresponding to the target energy device based on the correlation between the operating condition characteristic value of the target energy device and the parameter;
[0205] A sample data generating module 33 is used to determine sample data similar to the fitting curve with fitting error based on the deleted fitting curve with fitting error and in combination with existing samples;
[0206] The second curve fitting module 34 is used to perform curve fitting based on the sample data close to the fitting curve with fitting error and the sample data corresponding to the undeleted fitting curve in the curve cluster to obtain a curve cluster representing the device characteristics of the target energy device.
[0207] In this exemplary embodiment, the data characterizing the working condition of the target energy device includes working condition characteristic values and parameters corresponding to the working condition characteristic values. The working condition characteristic values and parameters corresponding to the working condition characteristic values are used to characterize various working conditions of the target energy device.
[0208] In this exemplary embodiment, the correlation between the operating characteristic value of the target energy device and the parameter includes a positive correlation or a negative correlation. If the correlation between the operating characteristic value of the target energy device and the parameter within the parameter interval is positively correlated and the positive correlation value is 1, or if the correlation between the operating characteristic value of the target energy device and the parameter is negatively correlated and the negative correlation value is -1, then theoretically the two characteristic curves will not intersect, and at this time, the curve generated in the previous step can be cross-checked to eliminate abnormal curves.
[0209] In this exemplary embodiment, the target energy equipment may include a chiller, an air source heat pump, a water source heat pump, a hot water boiler, a steam boiler, and the like. When the target energy equipment is a chiller, the operating condition characteristic values of the target energy equipment may include COP and refrigeration load, and the parameters of the target energy equipment include chilled water flow, cooling water flow, chilled water inlet temperature, and cooling water inlet temperature; when the target energy equipment is an air source heat pump, the operating condition characteristic values of the target energy equipment may include COP, refrigeration load / heating load, and the parameters of the target energy equipment include circulating water flow, inlet water temperature, and ambient temperature; when the target energy equipment is a water-ground source heat pump, the operating condition characteristic values of the target energy equipment may include COP, refrigeration load / heating load, and the parameters of the target energy equipment include user-side water flow, source-side water flow, user-side inlet water temperature, and source-side inlet water temperature; when the target energy equipment is a hot water boiler, the operating condition characteristic values of the target energy equipment may include thermal efficiency and heating load, and the parameters of the target energy equipment include circulating water flow and inlet water temperature; when the target energy equipment is a steam boiler, the operating condition characteristic values of the target energy equipment may include steam unit consumption and steam load, and the parameters of the target energy equipment include outlet steam pressure. Among them, the operating condition characteristic values corresponding to different target energy equipment may be different, and the parameters corresponding to different target energy equipment may also be different.
[0210] According to the energy equipment characteristic model construction device of the embodiment of the present disclosure, it is used to obtain sample data when the characteristic model of the target energy equipment is constructed; wherein the sample data is obtained by combining the parameter interval data corresponding to the parameters of the target energy equipment; based on the sample data when the characteristic model of the target energy equipment is constructed, curve fitting is performed to obtain a curve cluster corresponding to the target energy equipment; wherein each curve corresponds to the equipment characteristics of the target energy equipment under a working condition; based on the correlation between the working condition characteristic value and the parameter of the target energy equipment, the fitting curve with fitting error is deleted in the curve cluster corresponding to the target energy equipment; based on the deleted fitting curve with fitting error, combined with the existing samples, the sample data close to the fitting curve with fitting error is determined; based on the sample data close to the fitting curve with fitting error and the sample data corresponding to the undeleted fitting curve in the curve cluster, curve fitting is performed to obtain a curve cluster characterizing the equipment characteristics of the target energy equipment. In this application, curve fitting is performed on sample data corresponding to the parameters of the target energy device to obtain a curve cluster corresponding to the target energy device, and then the curve cluster is corrected, and sample data close to the fitting curve with fitting error is determined based on the deleted fitting curve and combined with the existing samples, and sample integration is performed to improve the validity of the sample data. Then, curve fitting is performed on the integrated sample data to obtain a curve cluster that can characterize the device characteristics of the target energy device.
[0211] In some embodiments, the data acquisition module 30 is used to
[0212] Obtain the parameter value range when constructing the characteristic model of the target energy equipment;
[0213] Segmenting the parameter value range by characteristic center to obtain N1 parameter interval data, wherein each parameter interval has a characteristic center;
[0214] The parameter interval data corresponding to all the parameters of the target energy device are combined to obtain sample data when the characteristic model of the target energy device is constructed.
[0215] In this exemplary embodiment, the sample data is obtained by combining the parameter interval data corresponding to the parameters of the target energy device. When obtaining the sample data when the characteristic model of the target energy device is constructed, the parameter value range when the characteristic model of the target energy device is constructed can be obtained; the parameter value range is segmented by the characteristic center to obtain N1 parameter interval data; each parameter interval has a characteristic center; the parameter interval data corresponding to all the parameters of the target energy device are combined to obtain the sample data when the characteristic model of the target energy device is constructed. Figure 3 As shown in the figure, the curve center initialization process in the process of building the energy equipment characteristic model includes:
[0216] Step 30, start;
[0217] Step 31, calculate the maximum value Max_i and the minimum value Min_i (i=1, 2, 3...N) of each dimension feature;
[0218] Step 32, according to the number of feature centers in each dimension N1, the data segment from the minimum value Min_i to the maximum value Max_i is divided into N1 segments, and the endpoints P_ij of each segment are obtained; where (i=1, 2, 3...N, j=1, 2, 3...N1);
[0219] Step 33, according to the data endpoints P_ij of each feature, the features are combined with each other to obtain the center vector of the ∏N1 curve, referred to as C_k (k=1, 2, 3...∏N1);
[0220] Step 34, end.
[0221] In this exemplary embodiment, for example, the maximum value and minimum value corresponding to each dimension of parameter data are obtained; according to the number of characteristic centers N1 of each dimension of parameter data, each dimension of parameter data is equally divided into data segments to obtain each parameter interval data including the endpoints of each segment of data; according to each parameter interval data of each dimension of data, the characteristics are combined with each other in units to obtain a parameter interval data combination including N1 to the power of N1; wherein, a parameter interval data combination is a curve center vector; and a curve center vector corresponds to a fitting curve. The parameter interval data combination of N1 to the power of N1 is the sample data for constructing the characteristic model of the target energy device.
[0222] The above samples are sample data based on two-dimensional parameter data. The number of parameter interval data combinations in the sample data is N1 to the power of N1; when constructing samples based on three-dimensional parameter data, if each parameter interval is N1 data interval, the number of parameter interval data combinations contained in the constructed sample data is N1 to the power of N1, and then N1 again.
[0223] In some embodiments, the first curve fitting module 31 is used to
[0224] Determine the characteristic interval corresponding to each parameter;
[0225] Determining fitting data for curve fitting in each parameter interval based on the characteristic center corresponding to each parameter interval and the characteristic interval corresponding to each parameter;
[0226] Based on the fitting data used for curve fitting in all parameter intervals, curve fitting is performed to obtain a curve cluster corresponding to the target energy equipment.
[0227] In this exemplary embodiment, when performing curve fitting through various parameter interval data combinations in the sample data, data screening can be performed on various parameter interval data combinations first to screen out interval valid data, and then curve fitting can be performed based on the interval valid data.
[0228] In this exemplary embodiment, Figure 4 As shown in the figure, the preliminary curve fitting process in the process of building the energy equipment characteristic model includes:
[0229] Step 40, start;
[0230] Step 41, calculate the maximum value Max_i and the minimum value Min_i (i=1, 2, 3...N) of each dimension feature;
[0231] Step 42, calculating the feature interval D1_i of each dimension;
[0232] Step 43: According to the curve center vector C_ki and the characteristic interval D1_i, sample data within the range of C_ki±D1_i / 2 is selected. At this time, the data only includes the load rate and the target characteristic value. The formula y=ax 2 +bx+c are fitted to obtain the undetermined parameters a_k, b_k, c_k (k=1, 2, 3...∏N1_i) corresponding to the center of each curve; among which, the load rate and the target characteristic value are both characteristic values of the working condition;
[0233] Step 44, end.
[0234] Among them, the characteristic intervals corresponding to the parameters include:
[0235] D1_i=Minimum(Maximum((Max_i-Min_i) / CC1_i, CC2_i), CC3_i); wherein D1_i is the characteristic interval corresponding to the i-th parameter; CC1_i is the curve center interval initialization factor corresponding to the i-th parameter; CC2_i is the lower limit of the curve center interval corresponding to the i-th parameter; CC3_i is the upper limit of the curve center interval corresponding to the i-th parameter; CC1_i=(Max_i-Min_i) / N1; CC2_i=CC1_i / 2; CC3_i=2*CC1_i;
[0236] The fitting data used for curve fitting in each parameter interval is (P_1n±D1_i / 2, P_2m±D1_j / 2); wherein (P_1n, P_2m) is the characteristic center corresponding to each parameter interval;
[0237] The step of determining fitting data for curve fitting in each parameter interval based on the characteristic center corresponding to each parameter interval and the characteristic interval corresponding to each parameter includes:
[0238] Based on the characteristic center (P_1n, P_2m) corresponding to each parameter interval and the characteristic interval D1_i corresponding to each parameter, the fitting data (P_1n±D1_i / 2, P_2m±D1_j / 2) used for curve fitting in each parameter interval is determined.
[0239] In this exemplary embodiment, Maximum9() is the maximum value between the two, and Minimum() is the minimum value between the two. After obtaining the interval valid data in the parameter interval data combination corresponding to each parameter, curve fitting can be performed based on the interval valid data in the parameter interval data combination corresponding to each parameter to obtain the curve cluster corresponding to the target energy device. If it is a curve fitting corresponding to a two-dimensional parameter, a two-dimensional curve fitting can be performed to obtain each fitting curve y=ax 2 +bx+c corresponding to the unknown parameters a, b, c, thus obtaining the fitting curve.
[0240] In some embodiments, based on the correlation between the operating condition characteristic value of the target energy device and the parameter, deleting the fitting curve with fitting error in the curve cluster corresponding to the target energy device includes:
[0241] Determining the theoretical intersection of curves between the fitting curves based on the correlation between the operating characteristic value of the target energy equipment and the parameters within the parameter intervals;
[0242] Based on the theoretical intersection of the fitting curves and the actual intersection of the fitting curves, the fitting curves with fitting errors are deleted from the curve cluster corresponding to the target energy equipment.
[0243] In this exemplary embodiment, based on the correlation between the operating condition characteristic value of the target energy equipment and the parameters in each parameter interval, determining the curve theoretical intersection between each fitting curve includes:
[0244] If the correlation between the operating characteristic value of the target energy equipment and the parameters in the two parameter intervals is positively correlated, and the positive correlation value is 1, or the correlation between the operating characteristic value of the target energy equipment and the parameters in each parameter interval is negatively correlated, and the negative correlation value is -1, then it can be determined that the fitting curves corresponding to the two parameter intervals are theoretically non-intersecting, otherwise the fitting curves corresponding to the two parameter intervals may intersect in theory. When the two fitting curves are theoretically non-intersecting, but the fitting curves obtained in the actual fitting process intersect, it means that at least one of the two fitting curves has a fitting error.
[0245] The fitting curve with fitting error can be determined by the intersection of a fitting curve and the other fitting curves, and then the fitting curve with fitting error is deleted.
[0246] The operating characteristic values of the target energy equipment include load rate and COP performance coefficient;
[0247] The determining of the curve theoretical intersection between each fitting curve based on the correlation between the operating condition characteristic value of the target energy equipment and the parameters within each parameter interval includes:
[0248] Based on the correlation between the COP performance coefficient and the parameters at the same load rate within each parameter interval, the curve theoretical intersection between each fitting curve is determined.
[0249] The determining of the theoretical intersection of the curves between the fitting curves based on the correlation between the COP performance coefficient and the parameters at the same load rate within each parameter interval includes:
[0250] Determine the curve center vector (P_1A, P_2A) of curve A, the curve center vector (P_1B, P_2B) of curve B, the chilled water flow characteristic correlation = DD3_1, and the cooling water flow characteristic correlation = DD3_2; wherein the chilled water flow characteristic correlation characterizes the correlation between the COP performance coefficient and the chilled water flow; the cooling water flow characteristic correlation characterizes the correlation between the COP performance coefficient and the cooling water flow; P_1A and P_2A are both characteristic centers of the parameters of curve A, and P_1B and P_2B are characteristic centers of the parameters of curve B; for example, P_1A is the characteristic center of the parameter chilled water flow of curve A, P_2A is the characteristic center of the parameter cooling water flow of curve A, P_1B is the characteristic center of the parameter chilled water flow of curve B, and P_2B is the characteristic center of the parameter cooling water flow of curve B;
[0251] Determine F1 = (P_1A-P_1B)*DD3_1, F2 = (P_2A-P_2B)*DD3_2;
[0252] Determine whether F1 is equal to F2;
[0253] If F1 is equal to F2, it means that the curves are consistent and theoretically there is no intersection between the curves;
[0254] If F1 is not equal to F2, it means that the curves are inconsistent and the curves can theoretically intersect.
[0255] like Figure 5 As shown in the figure, the process of correcting the fitting curve in the process of building the energy equipment characteristic model includes:
[0256] Step 50, start;
[0257] Step 51, counting the number of curves n1;
[0258] Step 52, traverse each curve, count the number of curves among the remaining curves that intersect the target curve in the load rate [DD1, DD2] interval and are consistent with the target curve, and obtain n_i (i=1, 2, 3...N);
[0259] Step 53: determine whether max(n_i) is greater than 0;
[0260] Step 54, if max(n_i)>0, delete the curve with the largest number of intersections, and then loop through steps 51 to 54;
[0261] Step 55: If max(n_i) is not greater than 0, end.
[0262] like Figure 6 As shown, after curve correction, we can get Figure 6 The curve cluster shown is after deleting the fitting curve with fitting error. Figure 6 The horizontal axis is the load factor and the vertical axis is the COP.
[0263] In some embodiments, the sample data generation module 33 is used to
[0264] Determine two sample curves that are closest to the characteristic center of the fitting curve with fitting error among the existing samples;
[0265] Based on the operating condition characteristic value of the target energy equipment corresponding to the fitting curve with fitting error, parameter sampling is performed on the two sample curves to obtain parameter sample data of the operating condition characteristic value corresponding to the fitting curve with fitting error;
[0266] Performing curve fitting based on parameter sample data of operating condition characteristic values corresponding to the fitting curve with fitting error to obtain a sample fitting curve;
[0267] The operating condition characteristic value of the target energy equipment corresponding to the fitting curve with the fitting error is substituted into the sample fitting curve to obtain sample data close to the fitting curve with the fitting error.
[0268] In this exemplary embodiment, since the sample data corresponding to the fitting curve with fitting error is also part of the overall sample data, the sample data corresponding to the deleted fitting curve can be reused. For example, in the existing samples, the two sample curves closest to the feature center of the fitting curve with fitting error are determined;
[0269] Based on the operating condition characteristic value of the target energy equipment corresponding to the fitting curve with fitting error, parameter sampling is performed on the two sample curves to obtain parameter sample data of the operating condition characteristic value corresponding to the fitting curve with fitting error;
[0270] Performing curve fitting based on parameter sample data of operating condition characteristic values corresponding to the fitting curve with fitting error to obtain a sample fitting curve;
[0271] The operating condition characteristic value of the target energy equipment corresponding to the fitting curve with the fitting error is substituted into the sample fitting curve to obtain sample data close to the fitting curve with the fitting error.
[0272] In this exemplary embodiment, Figure 7 As shown in the figure, the synthesis curve process in the process of building the energy equipment characteristic model includes:
[0273] Step 70, start;
[0274] Step 71, counting the curve samples deleted due to correction, there are M curves in total, each curve has n_k samples, (k=1, 2, 3...N);
[0275] Step 72, let i=1;
[0276] Step 73: Based on the characteristic center of the i-th curve, find two curves with the smallest Euclidean distance from its center, which are curve A and curve B;
[0277] Step 74: randomly collect EE1 load rates x_j (j=1, 2, 3...EE1) simultaneously in the load rate interval [DD1, DD2];
[0278] Step 75: According to x_j, collect corresponding target values y_A_j and y_B_j from curve A and curve B respectively;
[0279] Step 76: According to the N-dimensional eigenvalue, y_A_j is used as the target value, and the center of curve A is used as the N-dimensional independent variable to construct a sample. Similarly, a sample is constructed for curve B. According to the two samples, a linear equation of N variables is used for fitting to obtain fitting coefficients F_i (i=1, 2, 3...N);
[0280] Step 77, according to the fitting coefficient, calculate the target value when the independent variable is the characteristic center of the i-th curve, obtain the sample (x_j, y_j), and merge it with the original sample of the curve;
[0281] Step 78: Use the formula y=ax 2 +bx+c to fit and obtain the unknown parameters a_i, b_i, c_i corresponding to the center of each curve;
[0282] Step 79, determine whether i is greater than M;
[0283] Step 80: If i is not greater than M, then i=i+1, and steps 72 to 79 are executed in a loop;
[0284] Step 81: If i is greater than M, then end.
[0285] Specifically, after the correction of the fitting curve is completed, some fitting failure samples are obtained. For example, in the range of the center of the curve where the chilled water flow rate is 100 and the cooling water flow rate is 150, there are some fitting failure samples. Now it is necessary to use the fitted curve to use these samples.
[0286] First, among the generated curves, find the curve whose center is closest to (100,150), for example, find curve A whose center is (110,160) and curve B whose center is (90,140).
[0287] Secondly, we need to sample curves A and B by traversing the load rate, and then perform linear regression to generate samples. For example, we need to sample samples with a load rate of 40. We get sample COP=6 from curve A and sample COP=5 from curve B. According to the center of the corresponding curve, we assemble A: COP=6, chilled water flow=110, cooling water flow=160; B: COP=5, chilled water flow=90, cooling water flow=140.
[0288] Then, it is necessary to use COP as the target value and the center of the curve as the independent variable for linear regression fitting. The fitting formula is: y=k1·x1+k2·x2. The above two samples are used to fit the formula, and the parameters k1 and k2 to be determined are determined. Then, the center of the target curve x1=100, x2=150 is substituted, and COP=4 can be obtained. At this time, COP=4 and load rate=40 become the sample with the center of the curve (100,150). In this way, samples under different load rates can be obtained. Finally, the characteristic curve under the center of the curve can be fitted based on these samples.
[0289] In some embodiments, based on the theoretical intersection of the curves between the fitting curves and the actual intersection of the fitting curves, deleting the fitting curve with the fitting error in the curve cluster corresponding to the target energy device includes:
[0290] Among all the curve intersections, the curve that crosses the same curve the maximum number of times is determined as the target curve;
[0291] The target curve is determined to be a fitting curve with a fitting error, and is deleted.
[0292] In this exemplary embodiment, when determining a fitting curve with a fitting error, it is possible to determine the fitting curves among all the fitting curves that should not intersect with the same fitting curve, that is, the fitting curves with a positive correlation value of 1 or a negative correlation value of -1; then, the number of curves that actually intersect among the fitting curves with a positive correlation value of 1 or a negative correlation value of -1 is determined; among the number of curves that should not intersect but do intersect for each fitting curve, the fitting curve with the largest number of intersecting curves is the fitting curve with a fitting error; the fitting curve with a fitting error can be deleted.
[0293] The present disclosure provides a computer-readable storage medium on which an energy device characteristic model construction program is stored. When the energy device characteristic model construction program is executed by a processor, the energy device characteristic model construction method described in the above embodiments is implemented.
[0294] The present disclosure provides a server, including a memory, a processor, and an energy device characteristic model construction program stored in the memory and executable on the processor. When the processor executes the energy device characteristic model construction program, the energy device characteristic model construction method described in the above embodiments is implemented.
[0295] It should be noted that the logic and / or steps represented in the flowchart or otherwise described herein, for example, can be considered as a sequenced list of executable instructions for implementing logical functions, and can be specifically implemented in any computer-readable medium for use by an instruction execution system, device or apparatus (such as a computer-based system, a system including a processor, or other system that can fetch instructions from an instruction execution system, device or apparatus and execute instructions), or in combination with these instruction execution systems, devices or apparatuses. For the purposes of this specification, "computer-readable medium" can be any device that can contain, store, communicate, propagate or transmit a program for use by an instruction execution system, device or apparatus, or in combination with these instruction execution systems, devices or apparatuses. More specific examples of computer-readable media (a non-exhaustive list) include the following: an electrical connection portion with one or more wirings (electronic device), a portable computer disk box (magnetic device), a random access memory (RAM), a read-only memory (ROM), an erasable and programmable read-only memory (EPROM or flash memory), an optical fiber device, and a portable compact disk read-only memory (CDROM). In addition, the computer-readable medium may even be paper or other suitable medium on which the program is printed, since the program may be obtained electronically, for example, by optically scanning the paper or other medium and then editing, interpreting or otherwise processing in a suitable manner if necessary, and then stored in a computer memory.
[0296] It should be understood that the various parts of the present disclosure can be implemented in hardware, software, firmware or a combination thereof. In the above-mentioned embodiments, multiple steps or methods can be implemented in software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented in hardware, as in another embodiment, it can be implemented by any one of the following technologies known in the art or their combination: a discrete logic circuit having a logic gate circuit for implementing a logic function for a data signal, a dedicated integrated circuit having a suitable combination of logic gate circuits, a programmable gate array (PGA), a field programmable gate array (FPGA), etc.
[0297] In the description of this specification, the description with reference to the terms "one embodiment", "some embodiments", "example", "specific example", or "some examples" means that the specific features, structures, materials or characteristics described in conjunction with the embodiment or example are included in at least one embodiment or example of the present disclosure. In this specification, the schematic representation of the above terms does not necessarily refer to the same embodiment or example. Moreover, the specific features, structures, materials or characteristics described may be combined in any one or more embodiments or examples in a suitable manner.
[0298] In the description of the present disclosure, it should be understood that the terms "center", "longitudinal", "lateral", "length", "width", "thickness", "up", "down", "front", "back", "left", "right", "vertical", "horizontal", "top", "bottom", "inside", "outside", "clockwise", "counterclockwise", "axial", "radial", "circumferential" and the like indicate orientations or positional relationships based on the orientations or positional relationships shown in the accompanying drawings, and are only for the convenience of describing the present disclosure and simplifying the description, and do not indicate or imply that the referred device or element must have a specific orientation, be constructed and operated in a specific orientation, and therefore should not be understood as a limitation on the present disclosure.
[0299] In addition, the terms "first", "second", etc. used in the embodiments of the present disclosure are only used for descriptive purposes and should not be understood as indicating or implying relative importance, or implicitly indicating the number of technical features indicated in the embodiments. Therefore, the features defined by the terms "first", "second", etc. in the embodiments of the present disclosure may explicitly or implicitly indicate that at least one of the features is included in the embodiment. In the description of the present disclosure, the word "multiple" means at least two or two or more, such as two, three, four, etc., unless otherwise clearly and specifically defined in the embodiments.
[0300] In the present disclosure, unless otherwise clearly specified or limited in the embodiments, the terms "installed", "connected", "connected" and "fixed" etc. appearing in the embodiments should be understood in a broad sense. For example, the connection can be a fixed connection, a detachable connection, or an integrated connection. It can be understood that it can also be a mechanical connection, an electrical connection, etc.; of course, it can also be a direct connection, or an indirect connection through an intermediate medium, or it can be the internal connection of two elements, or the interaction relationship between two elements. For ordinary technicians in this field, the specific meanings of the above terms in the present disclosure can be understood according to the specific implementation situation.
[0301] In the present disclosure, unless otherwise clearly specified and limited, a first feature being "above" or "below" a second feature may mean that the first and second features are in direct contact, or the first and second features are in indirect contact through an intermediate medium. Moreover, a first feature being "above", "above" or "above" a second feature may mean that the first feature is directly above or obliquely above the second feature, or simply means that the first feature is higher in level than the second feature. A first feature being "below", "below" or "below" a second feature may mean that the first feature is directly below or obliquely below the second feature, or simply means that the first feature is lower in level than the second feature.
[0302] Although the embodiments of the present disclosure have been shown and described above, it is to be understood that the above embodiments are exemplary and are not to be construed as limitations of the present disclosure. A person skilled in the art may change, modify, replace and vary the above embodiments within the scope of the present disclosure.
Claims
1. A method for constructing an energy equipment characteristic model, characterized in that: include: Acquire sample data when constructing a characteristic model of a target energy device; wherein the sample data is obtained by combining parameter interval data corresponding to the parameters of the target energy device; Based on the sample data when the characteristic model of the target energy device is constructed, curve fitting is performed to obtain a cluster of curves corresponding to the target energy device; wherein each curve corresponds to the device characteristics of the target energy device under one working condition; Based on the correlation between the operating condition characteristic value of the target energy device and the parameter, deleting a fitting curve with a fitting error in the curve cluster corresponding to the target energy device; Based on the deleted fitting curve with fitting error, combined with existing samples, determine sample data close to the fitting curve with fitting error; Based on the sample data close to the fitting curve with fitting error and the sample data corresponding to the undeleted fitting curve in the curve cluster, curve fitting is performed to obtain a curve cluster characterizing the equipment characteristics of the target energy equipment.
2. The method for constructing an energy device characteristic model according to claim 1, characterized in that: The step of obtaining sample data when constructing a characteristic model of a target energy device includes: Obtain the parameter value range when constructing the characteristic model of the target energy equipment; Segmenting the parameter value range by characteristic center to obtain N1 parameter interval data, wherein each parameter interval has a characteristic center; The parameter interval data corresponding to all the parameters of the target energy device are combined to obtain sample data when the characteristic model of the target energy device is constructed.
3. The method for constructing an energy device characteristic model according to claim 1, characterized in that: The curve fitting is performed based on the sample data when the characteristic model of the target energy device is constructed to obtain a curve cluster corresponding to the target energy device, including: Determine the characteristic interval corresponding to each parameter; Determining fitting data for curve fitting in each parameter interval based on the characteristic center corresponding to each parameter interval and the characteristic interval corresponding to each parameter; Based on the fitting data used for curve fitting in all parameter intervals, curve fitting is performed to obtain a curve cluster corresponding to the target energy equipment.
4. The method for constructing an energy device characteristic model according to claim 3, characterized in that: The characteristic intervals corresponding to the parameters include: D1_i=Minimum(Maximum((Max_i-Min_i) / CC1_i, CC2_i), CC3_i); wherein D1_i is the characteristic interval corresponding to the ith parameter; CC1_i is the initialization factor of the curve center interval corresponding to the ith parameter; CC2_i is the lower limit of the curve center interval corresponding to the ith parameter; CC3_i is the upper limit of the curve center interval corresponding to the ith parameter; CC1_i=(Max_i-Min_i) / N1; CC2_i=CC1_i / 2; CC3_i=2*CC1_i; The fitting data used for curve fitting in each parameter interval is (P_1n±D1_i / 2, P_2m±D1_j / 2); wherein (P_1n, P_2m) is the characteristic center corresponding to each parameter interval; The step of determining fitting data for curve fitting in each parameter interval based on the characteristic center corresponding to each parameter interval and the characteristic interval corresponding to each parameter includes: Based on the characteristic center (P_1n, P_2m) corresponding to each parameter interval and the characteristic interval D1_i corresponding to each parameter , Determine the fitting data (P_1n±D1_i / 2, P_2m±D1_j / 2) for curve fitting in each parameter interval.
5. The method for constructing an energy device characteristic model according to claim 3, characterized in that: Based on the correlation between the operating condition characteristic value of the target energy device and the parameter, deleting a fitting curve with a fitting error in a curve cluster corresponding to the target energy device includes: Determining the theoretical intersection of curves between the fitting curves based on the correlation between the operating characteristic value of the target energy equipment and the parameters within the parameter intervals; Based on the theoretical intersection of the fitting curves and the actual intersection of the fitting curves, the fitting curves with fitting errors are deleted from the curve cluster corresponding to the target energy equipment.
6. The method for constructing an energy device characteristic model according to claim 1, characterized in that: The method of determining sample data close to the fitting curve with fitting error based on the deleted fitting curve with fitting error and combining it with existing samples includes: Determine two sample curves that are closest to the characteristic center of the fitting curve with fitting error among the existing samples; Based on the operating condition characteristic value of the target energy equipment corresponding to the fitting curve with fitting error, parameter sampling is performed on the two sample curves to obtain parameter sample data of the operating condition characteristic value corresponding to the fitting curve with fitting error; Performing curve fitting based on parameter sample data of operating condition characteristic values corresponding to the fitting curve with fitting error to obtain a sample fitting curve; The operating condition characteristic value of the target energy equipment corresponding to the fitting curve with the fitting error is substituted into the sample fitting curve to obtain sample data close to the fitting curve with the fitting error.
7. The method for constructing an energy device characteristic model according to claim 5, characterized in that: The method of deleting a fitting curve having a fitting error in a curve cluster corresponding to the target energy device based on the theoretical intersection of the fitting curves and the actual intersection of the fitting curves comprises: Among all the curve intersections, the curve that crosses the same curve the maximum number of times is determined as the target curve; The target curve is determined to be a fitting curve with a fitting error, and is deleted.
8. An energy equipment characteristic model building device, characterized in that: include: A data acquisition module, used to acquire sample data when constructing a characteristic model of a target energy device; wherein the sample data is obtained by combining parameter interval data corresponding to the parameters of the target energy device; A first curve fitting module is used to perform curve fitting based on sample data when the characteristic model of the target energy device is constructed, so as to obtain a curve cluster corresponding to the target energy device; wherein each curve corresponds to a device characteristic of the target energy device under one working condition; A curve deletion module, used for deleting a fitting curve with a fitting error in a curve cluster corresponding to the target energy device based on the correlation between the operating condition characteristic value of the target energy device and the parameter; A sample data generating module, used to determine sample data close to the fitting curve with fitting error based on the deleted fitting curve with fitting error and in combination with existing samples; The second curve fitting module is used to perform curve fitting based on the sample data close to the fitting curve with fitting error and the sample data corresponding to the undeleted fitting curve in the curve cluster to obtain a curve cluster characterizing the equipment characteristics of the target energy equipment.
9. A computer-readable storage medium, characterized in that: An energy device characteristic model building program is stored thereon, and when the energy device characteristic model building program is executed by a processor, the energy device characteristic model building method described in any one of claims 1 to 7 is implemented.
10. A server, characterized in that: The invention comprises a memory, a processor and an energy device characteristic model building program stored in the memory and executable on the processor. When the processor executes the energy device characteristic model building program, the energy device characteristic model building method described in any one of claims 1 to 7 is implemented.
Citation Information
Patent Citations
Acquiring system eigenfunction and signal feature value method
CN101158623A
Automatic inversion arithmetic based on height of laser radar mixed layer
CN101520515A
Wind turbine power curve modeling method
CN109002650A
Smart power grid monitoring data storage method based on discrete data curve fitting
CN113239030A
Method for automatically and accurately eliminating noisy points of detection data of white light interferometer
CN116222427A
