Non-intrusive household load state monitoring and identification method and system

Abstract

The invention discloses a non-intrusive household load state monitoring and identification method and system, and relates to the technical field of non-intrusive household load state monitoring and identification, and the method comprises the steps: judging whether a transient process exists in a period after a current time point or not based on a transient event monitoring criterion; determining a transient interval starting time point and load change through a sudden change point detection algorithm; and identifying a load state and an operation change by using an active power and current mark matching algorithm. According to the method, a transient process positioning mechanism can accurately determine the time interval of load start and stop; and the accuracy and adaptability of load state identification are further improved by combining a dynamic matching algorithm of active power and current imprint, so that the defects in a complex load scene in the prior art are effectively overcome.

Description

Technical Field

[0001] The present invention relates to the technical field of non-invasive household load status monitoring and identification, and specifically to a non-invasive household load status monitoring and identification method and system. Background Art

[0002] With the rapid development of smart grid and energy management technologies, non-intrusive load monitoring (NILM) technology has gradually become an important research direction in household energy management. By installing a single monitoring device at the power inlet, the NILM technology can achieve real-time monitoring and identification of household electricity load status without separately measuring household electrical appliances. Compared with traditional intrusive monitoring technologies, this method not only significantly reduces the hardware cost and installation complexity but also improves user acceptance. Currently, related research mainly focuses on steady-state load monitoring and transient load identification. Among them, steady-state monitoring focuses on the feature analysis of electrical equipment in a stable operating state, while transient identification focuses on sudden events of current or power fluctuations, and analyzes the start-stop behavior of the load status through algorithms. However, there is still much room for improvement in the response speed, accuracy, and adaptability to complex scenarios of existing transient monitoring technologies.

[0003] Existing non-intrusive load monitoring technologies, although able to achieve preliminary identification of steady-state and transient load status under specific conditions, still have obvious deficiencies in key technical indicators. First, most methods rely on steady-state features when detecting transient processes, ignoring the significance of short-time power fluctuation features, resulting in insufficient ability to capture rapid transient events, especially prone to misjudgment in complex scenarios where multiple devices are running simultaneously or frequently starting and stopping. Second, existing technologies lack a precise start-stop time positioning mechanism for transient interval identification, mostly relying on time segmentation methods with fixed windows, and unable to fully adapt to the complex changes in the load characteristics and operating states of different devices. In addition, existing load status identification algorithms are mostly limited to simple pattern matching and are difficult to effectively combine multi-dimensional electrical parameters (such as active power, reactive power, current, etc.) for comprehensive analysis, resulting in limited identification accuracy. Summary of the Invention

[0004] In view of the above problems, the present invention is proposed.

[0005] Therefore, the technical problem solved by the present invention is: existing non-intrusive load monitoring methods have insufficient ability to capture rapid transient events, inaccurate positioning of the start-stop time interval of the transient process, limited accuracy of load status identification, and how to achieve accurate and efficient load status monitoring and identification in complex load scenarios.

[0006] To solve the above technical problems, the present invention provides the following technical solution: a non-intrusive household load status monitoring and identification method, including judging whether there is a transient process in the next cycle after the current time point based on a transient event monitoring criterion; determining the starting time point of the transient interval and the load change through a mutation point detection algorithm; and identifying the load status and operation change by using an active power and current signature matching algorithm.

[0007] As a preferred embodiment of the non-intrusive household load status monitoring and identification method described in the present invention, wherein: the step of judging whether there is a transient process in the next cycle after the current time point includes screening the maximum power value P from N active power sequences corresponding to the next cycle T after the current time point K based on non-intrusive load monitoring. MAX and the minimum power value P MIN , and the active power sequence is expressed as:

[0008] p l ={p k+1 , p k+2 , …, p N+1}

[0009] The maximum power value is expressed as:

[0010] p MAX =MAX{p k+1 , p k+2 , …, p N+1}

[0011] The minimum power value is expressed as:

[0012] P MIN =MIN{p k+1 , p k+2 , …, p N+1}

[0013] Calculate the difference between P MAX and P MIN , which is expressed as:

[0014] Δp = P MAX - P MIN

[0015] where Δp represents the difference between P MAX and P MIN .

[0016] As a preferred embodiment of the non-intrusive household load status monitoring and identification method described in the present invention, wherein: the step of judging whether there is a transient process in the next cycle after the current time point further includes finding the minimum fluctuation amplitude of the active power of the transient process from the load signature library as a threshold, denoted as P H , and when Δp ≥ p HWhen there may be a transient process during the determination period T; when ΔP < p H When there is no transient process during the determination period T, the current time point is pushed back by one period T, and the determination of the next period is started.

[0017] As a preferred solution of the non-intrusive household load status monitoring and identification method described in the present invention, wherein: the starting time point of the transient interval includes when Δp ≥ p H When, calculate the average value of N points within the previous period T of the current time point, denoted as Expressed as:

[0018]

[0019] Respectively, denote the minimum active power point as P1 and the maximum power point as P2, and take K on both the left and right sides of P2, 3 ≤ K ≤ 0.25*N, and K is rounded. Calculate the absolute value of the difference between the neighborhoods of P1 and P2 and Then, find the minimum active power fluctuation amplitude of the transient process from the load imprint library as the threshold P M , and count the number of absolute values greater than P M , denoted as N1 and N2 respectively, and compare with the set threshold NS. If at least one of N1 and N2 is greater than NS, it is determined that there is a transient with a change in the load state after the current time point. If both N1 and N2 are less than NS, it is determined that there is no transient process during the determination period T, and the current time point is pushed back by one period T, and the determination of the next period is started.

[0020] As a preferred solution of the non-intrusive household load status monitoring and identification method described in the present invention, wherein: the load change includes determining the entry into the transient process, and is judged in four scenarios; Scenario 1 includes if N1 > NS and N2 ≤ NS, it is determined that there is a load input, or there is a load entering a higher power state from a lower power state; Scenario 2 includes if N2 > NS and N1 ≤ NS, it is determined that there is a load cut-off, or there is a load entering a lower power state from a higher power state; Scenario 3 includes if N1 > NS and N2 > NS, and the maximum power point is closer in time to the minimum power point, it is determined that there is a load input first or there is a load entering a higher power state from a lower power state, and then there is a load cut-off or there is a load entering a lower power state from a higher power state; Scenario 4 includes if N1 > NS and N2 > NS, and the minimum power point is closer in time to the maximum power point, it is determined that there is a load cut-off or there is a load entering a lower power state from a higher power state, and then there is a load input or there is a load entering a higher power state from a lower power state.

[0021] As a preferred solution of the non-invasive home load status monitoring and identification method described in the present invention, wherein: the identification of load status and operation changes includes, for the combination of Scenario 1 and Scenario 3, setting an active power threshold P F , and taking the point where the first power value greater than appears from the current point to before the maximum power as the starting point of the transient process; for the combination of Scenario 2 and Scenario 4, taking the point where the first power value less than appears from the current point to before the maximum power as the starting point of the transient process; for the combination of Scenario 1 and Scenario 2, starting from the Kth point after the maximum value, taking the absolute value of the difference of the active power until the number of consecutive absolute values of the difference less than the set active power threshold p H > Q × NS, Q > 1, then taking the leftmost point of the corresponding points as the end point of the transient process; for the combination of Scenario 3 and Scenario 4, dividing it into two consecutive transient processes and making judgments according to the method of the combination of Scenario 1 and Scenario 2, and finally determining two consecutive transient processes; performing harmonic detection on the current of the transient process in Scenario 1 based on the combination of ESMD and HT, and according to the obtained instantaneous amplitude and instantaneous frequency information of each harmonic, preliminarily matching the candidate load corresponding to the load switching and the state change.

[0022] As a preferred solution of the non-invasive home load status monitoring and identification method described in the present invention, wherein: the identification of load status and operation changes further includes calculating the similarity between the power sequence corresponding to the candidate load and state and the transient power sequence based on the DTW algorithm, and determining the candidate load and state with the highest similarity as the load and state of the transient process; the load and state in the steady-state stage of the latter stage after the transient process are obtained by superimposing the load and state in the steady-state stage before the transient process and the load and state of the transient process, or determined by superimposing and matching the current, active power, and reactive power of the load and state; circularly detecting and identifying based on the transient process, and determining the load and state of the adjacent next steady-state process by superimposition, so as to monitor and identify the evolution of the daily load and state in real time.

[0023] Another object of the present invention is to provide a non-invasive home load status monitoring and identification system, which can determine the starting time point of the transient interval and the load change through the mutation point detection algorithm, and solves the problem of low reliability in the current non-invasive load monitoring technology.

[0024] As a preferred solution of the non-invasive home load status monitoring and identification system described in the present invention, wherein: it includes a transient event judgment module, a transient interval determination module, and a load status identification module;

[0025] The transient event judgment module is used to judge whether there is a transient process in the next cycle after the current time point based on the transient event monitoring criterion; the transient interval determination module is used to determine the starting time point of the transient interval and the load change through the mutation point detection algorithm; the load state identification module is used to identify the load state and operation change by using the active power and current signature matching algorithm.

[0026] A computer device includes a memory and a processor. The memory stores a computer program, and the processor executes the computer program to implement the steps of the non-intrusive household load state monitoring and identification method.

[0027] A computer-readable storage medium stores a computer program thereon. When the computer program is executed by a processor, it implements the steps of the non-intrusive household load state monitoring and identification method.

[0028] The beneficial effects of the present invention: The transient process positioning mechanism of the non-intrusive household load state monitoring and identification method provided by the present invention can accurately determine the time interval of load start and stop; combined with the dynamic matching algorithm of active power and current signature, it further improves the accuracy and adaptability of load state identification, thus effectively solving the deficiencies of the prior art in complex load scenarios. The present invention has achieved better effects in terms of reliability, efficiency and accuracy. Description of the Drawings

[0029] In order to more clearly illustrate the technical solutions of the embodiments of the present invention, the drawings required for description in the embodiments will be briefly introduced below. Obviously, the drawings in the following description are only some embodiments of the present invention. For those of ordinary skill in the art, other drawings can be obtained based on these drawings without creative efforts.

[0030] Figure 1 It is the overall flowchart of a non-intrusive household load state monitoring and identification method provided by the first embodiment of the present invention.

[0031] Figure 2 It is a schematic diagram of a non-intrusive household load state monitoring and identification method provided by the second embodiment of the present invention.

[0032] Figure 3 It is a diagram for monitoring the start of the transient process and determining the transient process interval of a non-intrusive household load state monitoring and identification method provided by the second embodiment of the present invention.

[0033] Figure 4 It is a schematic diagram of the transient start stage of load startup of a non-intrusive household load state monitoring and identification method provided by the second embodiment of the present invention.

[0034] Figure 5 ESMD algorithm flowchart of a non-intrusive household load status monitoring and identification method provided for the second embodiment of the present invention.

[0035] Figure 6 Overall flowchart of a non-intrusive household load status monitoring and identification system provided for the third embodiment of the present invention. Detailed implementation manners

[0036] To make the above objects, features, and advantages of the present invention more obvious and understandable, the following describes the detailed implementation manners of the present invention in conjunction with the accompanying drawings of the specification. Obviously, the described embodiments are part of the embodiments of the present invention, rather than all embodiments. All other embodiments obtained by those of ordinary skill in the art based on the embodiments of the present invention without creative efforts shall fall within the protection scope of the present invention.

[0037] Embodiment 1, referring to Figure 1 , which is an embodiment of the present invention, provides a non-intrusive household load status monitoring and identification method, including:

[0038] S1: Based on the transient event monitoring criterion, determine whether there is a transient process in the next cycle after the current time point.

[0039] Furthermore, determining whether there is a transient process in the next cycle after the current time point includes, based on non-intrusive load monitoring, screening the maximum power value P MAX and the minimum power value P MIN from the N active power sequences corresponding to the next cycle T after the current time point K. The active power sequence is expressed as:

[0040] p l ={p k+1 , p k+2 , …, p N+1}

[0041] The maximum power value is expressed as:

[0042] p MAX = MAX{p k+1 , p k+2 , …, p N+1}

[0043] The minimum power value is expressed as:

[0044] P MIN = MIN{p k+1 , p k+2 , …, p N+1}

[0045] Calculate P MAX and P MINThe difference is expressed as:

[0046] Δp = P MAX - P MIN

[0047] where Δp represents the difference between P MAX and P MIN .

[0048] It should be noted that determining whether there is a transient process in the next cycle after the current time point also includes finding the minimum fluctuation amplitude of the active power of the transient process from the load imprint library as a threshold, denoted as P H , when Δp ≥ p H , it is determined that a transient process may occur within the period T; when ΔP < p H , it is determined that no transient process occurs within the period T, and then the current time point is pushed back one period T to start the determination of the next cycle.

[0049] It should also be noted that based on the transient event monitoring criterion, determining whether there is a transient process in the next cycle after the current time point realizes the rapid capture of short-term power fluctuations, filters out possible steady-state processes through the difference between the maximum value and the minimum value, so as to focus on more important transient events; ensures comprehensive monitoring of all possible load change processes through the rolling period method; can significantly improve the monitoring efficiency of transient events, reduce the interference to steady-state data, and thus reduce the possibility of misjudgment; at the same time, this rolling window-based determination mechanism has high robustness and real-time performance, laying a foundation for the precise positioning of the subsequent transient interval.

[0050] S2: Determine the starting time point of the transient interval and the load change through the mutation point detection algorithm.

[0051] Furthermore, the starting time point of the transient interval includes when Δp ≥ p H , calculate the average value of N points in the previous cycle T of the current time point, denoted as which is expressed as:

[0052]

[0053] Respectively, mark the minimum value point of the active power as P1 and the maximum value point as P2. For K points on both the left and right sides of P2, 3 ≤ K ≤ 0.25 * N, and K is an integer. Calculate the absolute value of the difference between the neighborhoods of P1 and P2 and , then find the minimum fluctuation amplitude of the active power of the transient process from the load imprint library as the threshold P M , and count the absolute values greater than P MThe numbers are respectively denoted as N1 and N2, and compared with the set threshold NS. If at least one of N1 and N2 is greater than NS, it is determined that there is a transient state change after the current time point. If both N1 and N2 are less than or equal to NS, it is determined that there is no transient process within the period T. Then, the current time point is pushed back by one period T to start the determination of the next period.

[0054] It should be noted that the load change includes determining the entry into the transient process, which is judged in four scenarios. Scenario 1 includes that if N1>NS and N2≤NS, it is determined that there is a load input, or there is a load changing from a lower power state to a higher power state. Scenario 2 includes that if N2>NS and N1≤NS, it is determined that there is a load cut-off, or there is a load changing from a higher power state to a lower power state. Scenario 3 includes that if N1>NS and N2>NS, and the maximum value point is closer in time to the minimum value point, it is determined that there is a load input first or there is a load changing from a lower power state to a higher power state, and then there is a load cut-off or there is a load changing from a higher power state to a lower power state. Scenario 4 includes that if N1>NS and N2>NS, and the minimum value point is closer in time to the maximum value point, it is determined that there is a load cut-off or there is a load changing from a higher power state to a lower power state, and then there is a load input or there is a load changing from a lower power state to a higher power state.

[0055] It should also be noted that through the mutation point detection algorithm, the starting time point of the transient interval and the load change are determined; it can accurately locate the starting and ending time points of the transient interval, and through the statistical analysis of the power fluctuation characteristics, further refine the change scenarios of the load state (such as load input, cut-off, etc.); this process ensures the high-precision monitoring of transient events in complex load scenarios; through the mutation point detection algorithm, the rapid identification and accurate positioning of the transient interval are realized; compared with the traditional method, it no longer depends on a fixed window time period, but dynamically adjusts according to the actual load characteristics, thereby improving the adaptability to complex load change scenarios and avoiding missed detections and false detections.

[0056] S3: Use the active power and current signature matching algorithm to identify the load state and operation changes.

[0057] Furthermore, identifying the load state and operation changes includes, for the combination of Scenario 1 and Scenario 3, setting an active power threshold P F , and taking the point where the first power value greater than appears from the current point to before the maximum power as the starting point of the transient process; for the combination of Scenario 2 and Scenario 4, taking the point where the first power value less than The point is used as the starting point of the transient process; for the combination of Scenario 1 and Scenario 2, the absolute value of the difference in active power is calculated starting from the Kth point after the maximum value until the absolute values of consecutive differences are all less than the set active power threshold p H If the number of points > Q × NS and Q > 1, then the leftmost point of the corresponding points is used as the end point of the transient process; for the combination of Scenario 3 and Scenario 4, it is divided into two consecutive transient processes and judged according to the method of the combination of Scenario 1 and Scenario 2 to finally determine two consecutive transient processes; for the current of the transient process in Scenario 1, harmonic detection based on the combination of ESMD and HT is performed, and according to the obtained instantaneous harmonic amplitudes and instantaneous frequency information, the candidate loads and state changes corresponding to the load switching are preliminarily matched

[0058] It should be noted that identifying the load state and operation changes also includes calculating the similarity between the power sequences corresponding to the candidate loads and states and the transient power sequence based on the DTW algorithm, and determining the candidate loads and states with the highest similarity as the loads and states of the transient process; the loads and states in the steady-state stage of the latter stage of the transient process are obtained by superimposing the loads and states in the steady-state stage before the transient process and the loads and states of the transient process, or determined by superimposing and matching the current, active power, and reactive power of the loads and states; based on the transient process detection and identification in a loop, the loads and states of the next adjacent steady-state process are determined by superimposition, thereby monitoring and identifying the evolution of the daily load and state in real time

[0059] It should also be noted that by using the power and current data in the transient interval and through comprehensive matching with the imprint library, the operation state of the load is identified, including load connection, disconnection, and subtle changes in the load state; by superimposing the load state and the steady-state result, the operation evolution of the global load can be dynamically monitored; the accuracy of load identification is significantly improved through comprehensive matching of multi-dimensional features, especially in complex scenarios where multiple devices are operating simultaneously, and the state changes of different loads can be accurately distinguished; in addition, through the joint analysis of the steady state and the transient state, dynamic tracking of the load state is realized, providing accurate data support for energy management

[0060] Example 2, referring to Figures 2 - 5 , which is an embodiment of the present invention, provides a non-intrusive household load state monitoring and identification method. In order to verify the beneficial effects of the present invention, scientific demonstration is carried out through economic benefit calculation and simulation experiments

[0061] First, the experimental object is the power system of a household, mainly including various load devices such as refrigerators, electric water heaters, air conditioners, etc., and a smart meter is used for global monitoring Figure 2This is a schematic diagram of the method of the present invention. Using the transient event monitoring criterion, with a period of every 3 seconds, the active power sequence corresponding to the period is extracted, and the maximum power value and the minimum power value are screened; the difference between the two is calculated and compared with the threshold value to determine whether there is a transient process; when Δp≥p H , the mutation point detection algorithm is used to further analyze the power change characteristics, determine the starting time of the transient interval, and identify possible load start / stop events according to the load change law; referring to Figure 3 for the start monitoring of the transient process and the determination of the transient process interval, Figure 4 represents the transient stage of load start. Using the active power and current imprint matching algorithm, the detected load state is matched with the characteristics in the imprint library, so as to identify the change state of specific load equipment, such as load increase, decrease or compound change; Figure 5 represents the ESMD algorithm flow. Among them, x(t) is the input signal, that is, the time series or waveform data; R is the remaining signal, which represents the signal after continuously removing components after multiple iterations; L u is the upper envelope line, an interpolation curve constructed by local extreme points; L p is the lower envelope line, an interpolation curve constructed by local extreme points; L * is the average value curve, which represents the average trend of the upper and lower envelope lines; h is the current signal component, the new signal after removing the average value; n is the number of iterations, used to track the current processing progress; ∈ is the convergence condition, which represents the allowable error range of L * ; P is the number of envelope lines, usually two for the upper and lower envelope lines; j is the maximum number of iterations, used to prevent infinite loops; M is the extracted signal component, used to update the remaining signal.

[0062] During the experiment, record P MAX 、P MIN 、Δp and whether a transient event is detected, and further analyze the change of the load state.

[0063] In the period of 6 - 9 seconds, the system detects the load removal state, indicating that the mutation point detection algorithm can quickly respond to power fluctuations and accurately locate the transient interval; this ability significantly improves the adaptability to rapidly changing loads; data shows that in complex scenarios (such as 15 - 18 seconds), the system can detect the composite state of load increase and removal at the same time; this shows that through the active power and current imprint matching algorithm, multiple load states can be effectively distinguished, avoiding the problem of multi-device interference in the prior art; traditional technologies often rely on fixed time windows or single parameters, while the present invention combines multi-dimensional data analysis to improve the accuracy and adaptability of detection; in addition, through the dynamic adjustment method of the rolling window, the occurrence of missed detection of transient events is avoided.

[0064] Example 3, referring toFigure 6 , which is an embodiment of the present invention, provides a non-intrusive home load status monitoring and identification system, including a transient event judgment module, a transient interval determination module, and a load status identification module.

[0065] Among them, the transient event judgment module is used to judge whether there is a transient process in the next cycle after the current time point based on the transient event monitoring criterion; the transient interval determination module is used to determine the starting time point of the transient interval and the load change through the mutation point detection algorithm; the load status identification module is used to identify the load status and operation change by using the active power and current signature matching algorithm.

[0066] If the function is implemented in the form of a software functional unit and sold or used as an independent product, it can be stored in a computer-readable storage medium. Based on such an understanding, the technical solution of the present invention, in essence, or the part that contributes to the prior art, or a part of this technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions for causing a computer device (which can be a personal computer, a server, or a network device, etc.) to execute all or part of the steps of the methods in various embodiments of the present invention. The aforementioned storage medium includes: various media such as USB flash drives, mobile hard disks, read-only memory (ROM, Read-Only Memory), random access memory (RAM, Random Access Memory), magnetic disks, or optical discs that can store program codes.

[0067] The logic and / or steps represented in the flowchart or described in other ways herein, for example, can be considered as a definite sequence 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, apparatus, or device (such as a computer-based system, a system including a processor, or other systems that can fetch instructions from the instruction execution system, apparatus, or device and execute the instructions), or used in combination with these instruction execution systems, apparatus, or devices. For the purposes of this specification, a "computer-readable medium" can be any device that can contain, store, communicate, propagate, or transmit a program for use by or in combination with an instruction execution system, apparatus, or device.

[0068] More specific examples (a non-exhaustive list) of computer-readable media include the following: electrical connections (electronic devices) having one or more wirings, portable computer diskettes (magnetic devices), random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM or flash memory), optical fiber devices, and portable compact disc read-only memory (CDROM). Additionally, the computer-readable media can even be paper or other suitable media on which a program can be printed, because the program can be obtained electronically, for example, by optically scanning the paper or other media, followed by editing, interpretation, or otherwise processing as appropriate, and then storing it in a computer memory.

[0069] It should be understood that the various parts of the present invention can be implemented by hardware, software, firmware, or a combination thereof. In the above embodiments, multiple steps or methods can be implemented by software or firmware stored in a memory and executed by a suitable instruction execution system. For example, if implemented by hardware, as in another embodiment, any one or a combination of the following techniques well known in the art can be used: discrete logic circuits having logic gate circuits for implementing logical functions on data signals, application specific integrated circuits having appropriate combinational logic gate circuits, programmable gate arrays (PGA), field programmable gate arrays (FPGA), etc. It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

[0070] It should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention and not to limit them. Although the present invention has been described in detail with reference to the preferred embodiments, those of ordinary skill in the art should understand that the technical solutions of the present invention can be modified or equivalently replaced without departing from the spirit and scope of the technical solutions of the present invention, and they should all be covered by the scope of the claims of the present invention.

Claims

1. A non-intrusive household load status monitoring and identification method, characterized in that: include: Based on the transient event monitoring criteria, determine whether there is a transient process in the cycle after the current time point; Determine the starting time point of the transient interval and load changes through the mutation point detection algorithm; Use active power and current signature matching algorithms to identify load states and operating changes.

2. The non-intrusive household load status monitoring and identification method according to claim 1, characterized in that: The determination of whether there is a transient process in a period after the current time point includes selecting the maximum power value P from the N active power sequences corresponding to the period T after the current time point K based on non-intrusive load monitoring. MAX and the minimum power value P MIN , the active power sequence is expressed as: p l ={p k+1 ,p k+2 ,…,p N+1 } The maximum power value is expressed as: p MAX = MAX P{p k+1 ,p k+2 ,…,p N+1 } The minimum power value is expressed as: P MIN = MIN {p k+1 ,p k+2 ,…,p N+1 } Calculate P MAX and P MIN The difference is expressed as: Δp=P MAX -P MIN Where Δp represents P MAX and P MIN The difference.

3. The non-intrusive household load status monitoring and identification method according to claim 2, characterized in that: The method of determining whether there is a transient process in a period after the current time point also includes finding the minimum fluctuation amplitude of the active power in the transient process from the load imprint library as a threshold value, denoted as P H , when Δp ≥ p H When , it is determined that a transient process may occur within the period T; When ΔP <p H If no transient process occurs within the judgment period T, the current time point is pushed back by one period T to start the judgment of the next period.

4. The non-intrusive household load status monitoring and identification method according to claim 3, characterized in that: The transient interval starting time point includes when Δp≥p H When , calculate the average value of N points in the period T before the current time point, recorded as It is expressed as: The minimum active power point is recorded as P1 and the maximum active power point is recorded as P2. The K points around the points are 3≤K≤0.25*N, and K is rounded. Calculate the neighborhood of P1 and P2 and The absolute value of the difference is then used to find the minimum fluctuation amplitude of the active power in the transient process from the load imprint library as the threshold value P M , the absolute value of the statistic is greater than P M The number of them is recorded as N1 and N2 respectively, and compared with the set threshold NS. If at least one of the values ​​of N1 and N2 is greater than NS, it is determined that there is a transient state of load state change after the current time point. If N1 and N2 are both Ns, it is determined that no transient process occurs within the period T, then the current time point is pushed back by one period T, and the determination of the next period is started.

5. The non-intrusive household load status monitoring and identification method according to claim 4, characterized in that: The load change includes determining the entry into the transient process, which is divided into four scenarios; Scenario 1 includes if N1>NS, N2≤NS, it is determined that there is a load input, or there is a load entering a higher power state from a lower power state; Scenario 2 includes if N2>NS, N1≤NS, it is determined that there is load shedding, or there is load entering a lower power state from a higher power state; Scenario 3 includes if N1>NS, N2>NS, and the maximum value point is closer in time to the minimum value point, it is determined that the load is first put into operation or the load enters a higher power state from a lower power state, and then the load is removed or the load enters a lower power state from a higher power state; Scenario 4 includes if N1>NS, N2>NS, and the minimum point is closer in time to the maximum point, it is determined that the load is removed or the load enters a lower power state from a higher power state, and then the load is put into operation or the load enters a higher power state from a lower power state.

6. The non-intrusive household load status monitoring and identification method according to claim 5, characterized in that: The identification of load status and operation changes includes setting an active power threshold P for the combination of scenario 1 and scenario 3. F , move the current point to the first power value before the maximum power greater than The point is taken as the starting point of the transient process; For the combination of scenario 2 and scenario 4, the current point to the first point before the maximum power where the power value is less than The point is taken as the starting point of the transient process; For the combination of scenario 1 and scenario 2, the absolute value of the difference in active power is calculated starting from the Kth point after the maximum value, until the absolute value of the continuous difference is less than the set active power threshold p H If the number of > Q×NS, Q>1, the leftmost point of the corresponding point is taken as the end point of the transient process; The combination of scenario 3 and scenario 4 is divided into two consecutive transient processes and judged according to the combination method of scenario 1 and scenario 2, and finally two consecutive transient processes are determined; For the current of the transient process in scenario 1, harmonic detection based on the combination of ESMD and HT is performed. According to the instantaneous amplitude and instantaneous frequency information of each harmonic, the candidate loads and state changes corresponding to the load switching are preliminarily matched.

7. The non-intrusive household load status monitoring and identification method according to claim 6, characterized in that: The identifying load state and operation change also includes calculating the similarity between the power sequence corresponding to the candidate load and state and the transient power sequence based on the DTW algorithm, and determining the candidate load and state with the highest similarity as the load and state of the transient process; The load and state of the steady-state phase after the transient process are obtained by superimposing the load and state of the steady-state phase before the transient process on the load and state of the transient process, or by superimposing and matching the current, active power and reactive power of the load and state; The cycle is based on transient process detection and identification, and the load and state of the next adjacent steady-state process are determined by superposition, thereby monitoring and identifying the evolution of intraday load and state in real time.

8. A system using the non-intrusive household load status monitoring and identification method according to any one of claims 1 to 7, characterized in that: It includes transient event judgment module, transient interval determination module and load state identification module; The transient event judgment module is used to judge whether there is a transient process in a period after the current time point based on the transient event monitoring criterion; The transient interval determination module is used to determine the starting time point of the transient interval and the load change through a mutation point detection algorithm; The load state identification module is used to identify the load state and operation changes by using active power and current imprint matching algorithm.

9. A computer device comprising a memory and a processor, wherein the memory stores a computer program, wherein: When the processor executes the computer program, the steps of the non-intrusive household load status monitoring and identification method according to any one of claims 1 to 7 are implemented.

10. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the non-intrusive household load status monitoring and identification method according to any one of claims 1 to 7 are implemented.

Images