Fault model rapid matching method based on feature quantity extraction and multilayer hash storage

Abstract

The invention discloses a fault model rapid matching method based on feature quantity extraction and multilayer hash storage. According to the method, fault model formula calculation is carried out on point location data collected by a system, and all formulas of point locations contained in model formulas in a batch of point location set need to be quickly found out in a large number of fault models. The method comprises the following steps: establishing a point location combination calculation rule for all fault models, extracting feature quantities from model formulas, establishing multi-layer hash storage of the feature quantities and corresponding fault model formulas, constructing a point location dictionary, and mapping each point code into a unique integer; the method comprises the following steps of: obtaining a set in which integers mapped by point locations of uploading data are ranked from small to large, obtaining a union set of an integer set after the mapping of the acquisition point locations and a characteristic quantity set, carrying out first-layer filtering, and carrying out multi-layer filtering and screening through an established hash table to finally obtain the remaining model formulas needing to be calculated in the point location acquisition data of the batch.

Description

Technical Field

[0001] The present invention relates to the technical field of rapid matching of fault models, and particularly to a method for rapid matching of fault models based on feature quantity extraction and multi-layer hash storage. Background Art

[0002] In the field of intelligent operation and maintenance, the monitoring scope continues to expand, covering an increasing number of devices and scenarios. At the same time, the types of faults also show a trend of diversification and complexity. From common equipment component aging, short circuits, and overloads to faults caused by factors such as bad weather and external construction interference, there is a wide variety of types. Different fault types require different fault models for diagnosis and analysis, which makes the number of fault models increase continuously. In actual operation and maintenance, each fault model contains specific point code combination calculations for determining whether a fault has occurred and the specific type and location of the fault. This makes it a key problem faced by intelligent operation and maintenance to quickly and accurately match the corresponding fault model in a large amount of operation data. Summary of the Invention

[0003] A method for rapid matching of fault models based on feature quantity extraction and multi-layer hash storage proposed by the present invention can at least solve one of the technical problems in the background art.

[0004] To achieve the above object, the present invention adopts the following technical solutions:

[0005] A method for rapid matching of fault models based on feature quantity extraction and multi-layer hash storage includes the following steps:

[0006] S100. Encode the point positions of the fault model formula and extract feature quantities;

[0007] S200. Rapidly filter the feature quantities of the collected point positions through the fault model formula;

[0008] S300. Extract the filtered data to achieve effective fault model calculation.

[0009] Further, the method for encoding the point positions of the fault model formula and extracting feature quantities in step S100 of the present invention includes:

[0010] Let the set of fault model formulas be M = {M1, M2, M3,..., M n}, where M i represents the i-th fault model formula. For each fault model formula M i , its set of point position codes is P i = {p i1 , p i2 , p i3 ,... p im}, where pij Denote the j-th point encoding in the i-th fault model formula.

[0011] S110. Construct a dictionary to map each point encoding to a unique integer;

[0012] Let the set of point encodings be p = {p1, p2, p3,... p k}, and construct a mapping function f: P -> Z to map each point encoding p i to a unique integer f(p j ) = z j , where Z represents the set of integers;

[0013] S120. Extract the point encodings in each fault model formula, obtain the set of feature quantities after mapping for each fault model, and establish a hash storage for the hash value of the result and the fault model formula;

[0014] For each fault model formula M i , the set of feature quantities after its mapping is:

[0015] Fi = {f(p i1 ), f(p i2 ), f(p i3 ),..., f(p im )}

[0016] Calculate its hash value h(F i ), where the hash function is h: 2 Z -> H, and H represents the set of hash values. Using h(F i ) as the key value, establish a hash storage H1 with the set of fault model formulas having the same hash value as the key value, satisfying: H1(h(F i )) = {M j |h(F j ) = h(F i ), j = 1, 2,... n};

[0017] S130. Merge and deduplicate the sets of feature quantities in all fault models to obtain the full set of feature quantities A for the fault model formulas;

[0018] Let the full set of feature quantities The deduplication operation is expressed as A = {a|a ∈ Fi, i = 1, 2,... n}, where a is unique.

[0019] S140. For each element in the full set of feature quantities A, establish a hash storage for the set of hash values of the feature quantity sets of all fault model formulas containing this feature quantity;

[0020] For each element a ∈ A, find all the sets of fault model formulas containing this element

[0021] M a ={M j |a∈F j , j = 1, 2, ... n},

[0022] Calculate the hash value set H of its feature quantity set a ={h(F j )|M j ∈M a}}, with a as the key and H a as the key value to establish a hash storage H2, satisfying: H2(a)=H a ;

[0023] S150. Merge and deduplicate the hash values of the feature quantity sets in all fault models to obtain the full - quantity hash value set B of the fault model formula;

[0024] Let B={h(F i )|i = 1, 2, ... n}, and the deduplication operation is expressed as B={b|b = h(F i ), i = 1, 2, ..., n}, b is unique.

[0025] Furthermore, the hash storage method in step S120 of the present invention includes:

[0026] S121. Extract the feature quantity set after mapping of each fault model and calculate the hash value of the feature quantity set;

[0027] For the fault model formula M i , obtain the feature quantity set F i in the following way and calculate h(F i ), using the hash function h to satisfy:

[0028] where hash represents the hash calculation function, and SHA - 256 is selected;

[0029] S122. Establish a hash storage with the hash value of the feature quantity set as the key value and the model formula set with the same hash value for all feature quantity sets as the key value;

[0030] The H1 defined in step S120 above satisfies: H1(h(F i ))={M j |h(F j ) = h(F i ), j = 1, 2, ... n}.

[0031] Furthermore, the hash storage method in step S140 of the present invention includes:

[0032] For each element a ∈ A, find all sets of fault model formulas that contain this element

[0033] M a ={M j | a ∈ F j , j = 1, 2,... n},

[0034] Calculate the set of hash values H of its set of characteristic quantities a ={h(F j ) | M j ∈ M a}, and establish a hash storage H2 with a as the key and H a as the key value, satisfying: H2(a) = H a ;

[0035] S141: Extract each element of the full set of characteristic quantities;

[0036] For each element a ∈ A, where A is the full set of characteristic quantities;

[0037] S142: Find all fault models that contain this characteristic quantity;

[0038] Find the set M a ={M j | a ∈ F j , j = 1, 2,... n};

[0039] S143: Calculate the hash value of the set of characteristic quantities of all matching fault models;

[0040] For each M j ∈ M a , calculate the hash value h(F j ) of its set of characteristic quantities F j );

[0041] S144: Use each element of the full set of characteristic quantities as the key value and the set of hash values of the sets of characteristic quantities of all matching fault model formulas as the key value for hash storage.

[0042] Furthermore, the method for quickly filtering the characteristic quantities of the collected points in the present invention is:

[0043] S210: Sort the point codes collected according to the unique integers mapped by the constructed dictionary in ascending order to obtain the set C;

[0044] Let the set of point codes collected be P' = {p'1, p'2, p'3,... p' s} and the mapped set be C = {f(p'1), f(p'2), f(p'3),... f(p' s)}, and perform a sorting operation on C, satisfying:

[0045] C = sort({f(p' j ) | p' j ∈P'})

[0046] S220. Obtain the intersection D of the feature quantity set A in step S130 and the set C;

[0047] Set D = A ∩ C;

[0048] S230. Obtain the feature quantity set E after removing the intersection D from the feature quantity set A;

[0049] Set E = A - D;

[0050] S240. Look up the hash table established in step S144 according to the feature quantity set E to obtain all the hash value sets F that need to be filtered;

[0051] Among them, the set F is:

[0052] S250. Obtain the set H after removing the hash value set F from the hash value set B, that is, obtain the hash value set of the feature quantity set of all the model formulas that need to calculate the point positions;

[0053] Set H = B - F.

[0054] Furthermore, the method for extracting the effective fault model in step S300 of the present invention is:

[0055] S310. Traverse the set H;

[0056] S320. Quickly query the model formula through the hash table established in step 122;

[0057] Query the model formula set M through H1(h) h = H1(h);

[0058] S330. Perform data substitution calculation on the model formula;

[0059] For each M i ∈M h , substitute the collected data into the formula M i for calculation. Assume the collected data is D = {d1, d2, d3,... d t} and substitute the data into the fault model formula M i for fault diagnosis and calculation to judge the fault state.

[0060] As can be seen from the above technical solutions, the present invention realizes the rapid filtering, screening, and matching of fault models by constructing a multi-layer hash storage and filtering algorithm. It has significant advantages in aspects such as the matching efficiency of fault models, the consumption of computing resources, the accuracy and reliability of fault diagnosis, as well as the scalability and adaptability of the system, providing an efficient and reliable fault diagnosis and maintenance solution for the field of intelligent operation and maintenance, and having important practical value and broad application prospects. BRIEF DESCRIPTION OF THE DRAWINGS

[0061] Figure 1 is a flowchart of feature quantity extraction and filtering processing in an embodiment of the present invention;

[0062] Figure 2 is a schematic diagram of the multi-layer hash storage structure of the present invention;

[0063] Figure 3 is the overall flowchart of the present invention. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0064] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Apparently, the described embodiments are some, but not all, of the embodiments of the present invention.

[0065] As Figure 1 shown, the rapid fault model matching method based on feature quantity extraction and multi-layer hash storage described in this embodiment is executed by a computer device through the following steps.

[0066] S100. Encode the points of the fault model formula and extract feature quantities;

[0067] S200. Rapidly filter the feature quantities of the collected points through the fault model formula;

[0068] S300. Extract the filtered data to implement effective fault model calculation.

[0069] The following is a detailed description of each step:

[0070] S100. Encode the points of the fault model formula and extract feature quantities;

[0071] Let the set of fault model formulas be M = {M1, M2, M3,..., M n}, where M i represents the i-th fault model formula. For each fault model formula M i , its set of point encodings is P i = {p i1 , p i2 , p i3,...p im}, where p_i j represents the j-th point encoding in the i-th fault model formula.

[0072] S110. Construct a dictionary to map each point encoding to a unique integer;

[0073] Let the set of point encodings be p = {p1, p2, p3,... p k}, construct a mapping function f: P -> Z, which maps each point encoding p i to a unique integer f(p j ) = z j , where Z represents the set of integers.

[0074] S120. Extract the point encodings in each fault model formula, obtain the set of feature quantities after mapping for each fault model, and establish a hash storage for the hash value of the result and the fault model formula;

[0075] For each fault model formula M i , the set of feature quantities after its mapping is:

[0076] Fi = {f(p i1 ), f(p i2 ), f(p i3 ),..., f(p im )}

[0077] Calculate its hash value h(F i ), where the hash function is h: 2 Z -> H, and H represents the set of hash values. Using h(F i ) as the key value, establish a hash storage H1 with the set of fault model formulas containing the same hash value as the key value, satisfying:

[0078] H1(h(F i )) = {M j |h(F j ) = h(F i ), j = 1, 2,... n}.

[0079] S121. Extract the set of feature quantities after mapping for each fault model and calculate the hash value of the set of feature quantities; for the fault model formula M i , obtain the set of feature quantities F i in the above way and calculate h(F i ), using the hash function h to satisfy:

[0080] where hash represents the hash calculation function, and SHA-256 is selected.

[0081] S122. Establish a hash storage with the hash value of the feature quantity set as the key value and the set of model formulas with the same hash value for all feature quantity sets as the key value.

[0082] As defined by H1 in the above step S120, it satisfies: H1(h(F i )) = {M j |h(F j ) = h(F i ), j = 1, 2,... n}

[0083] S130. Merge and deduplicate the feature quantity sets in all fault models to obtain the full - quantity feature quantity set A of the fault model formulas;

[0084] Let the full - quantity feature quantity set The deduplication operation is expressed as A = {a|a ∈ Fi, i = 1, 2,... n}, where a is unique.

[0085] S140. For each element in the full - quantity feature quantity set A, establish a hash storage with the element as the key and the set of hash values of the feature quantity sets of all fault model formulas containing this feature quantity as the key value;

[0086] As Figure 2 shown, for each element a ∈ A, find all the fault model formula sets

[0087] M a = {M j |a ∈ F j , j = 1, 2,... n},

[0088] Calculate the set of hash values H a = {h(F j )|M j ∈ M a}, and establish a hash storage H2 with a as the key and H a as the key value, satisfying: H2(a) = H a ;

[0089] S141: Extract each element of the full - quantity feature quantity;

[0090] For each element a ∈ A, where A is the full - quantity feature quantity set.

[0091] S142: Find out all the fault models containing this feature quantity;

[0092] Find the set M a = {M j |a ∈ F j , j = 1, 2,... n}.

[0093] S143: Calculate the hash values for the sets of characteristic quantities of all matching fault models;

[0094] For each M j ∈M a , calculate the set of its characteristic quantities F j and the hash value h(F j ).

[0095] S144: Use each element of the full set of characteristic quantities as the key value and the set of hash values of the sets of characteristic quantities that all match the fault model formula as the key value for hash storage.

[0096] Establish the hash storage H2 in the above manner, satisfying: H2(a) = {h(F j ) | M j ∈M a}.

[0097] S150: Merge and deduplicate the hash values of the sets of characteristic quantities in all fault models to obtain the full set of hash values B for the fault model formula;

[0098] Let B = {h(F i ) | i = 1, 2,... n}, and the deduplication operation is expressed as B = {b | b = h(F i ), i = 1, 2,..., n}, where b is unique.

[0099] S200: Quickly filter the characteristic quantities of the collected points through the fault model formula;

[0100] The method for quickly filtering the characteristic quantities of the collected points is as follows:

[0101] S210: Sort the point codes collected in ascending order according to the unique integers mapped by the constructed dictionary to obtain the set C;

[0102] Let the set of point codes collected be P' = {p'1, p'2, p'3,... p' s}, and the mapped set be C = {f(p'1), f(p'2), f(p'3),... f(p' s )}, and perform a sorting operation on C, satisfying: C = sort({f(p' j ) | p' j ∈P'})

[0103] S220: Obtain the intersection D of the set of characteristic quantities A in step S130 and the set C;

[0104] The set D = A ∩ C.

[0105] S230: Obtain the set of characteristic quantities E after removing the intersection D from the set of characteristic quantities A;

[0106] The set E = A - D.

[0107] S240. Find all the hash value sets F to be filtered from the hash table established in step 144 according to the feature quantity set E;

[0108] Among them, the set F is:

[0109] S250. Obtain the set H by removing the hash value set F from the hash value set B, that is, obtain the hash value set of the feature quantity set of all the model formulas for which the points need to be calculated;

[0110] The set H = B - F.

[0111] S300. Extract the filtered data to implement effective fault model calculation.

[0112] The method for extracting and implementing the effective fault model is as follows:

[0113] S310. Traverse the set H;

[0114] S320. Quickly query the model formula through the hash table established in step 122;

[0115] Query the model formula set M through H1(h) h = H1(h).

[0116] S330. Substitute data into the model formula for calculation.

[0117] For each M i ∈M h , substitute the collected data into the formula M i for calculation. Suppose the collected data is D = {d1, d2, d3,... d t}, substitute the data into the fault model formula M i for fault diagnosis and calculation to judge the fault state.

[0118] As Figure 3 shown, the following is a specific description through four fault model formulas:

[0119] Step1:

[0120] Extract the feature quantities of the fault model formula and establish a multi-layer hash storage.

[0121] Step1.1:

[0122] Suppose a set of fault model formulas

[0123] M1: p1 + 2p2 - 3p3 = 0

[0124] M2: p2 + p3 = 5

[0125] M3: 4p1 - p2 = 10

[0126] M4 = p1 + 5p2 = 6

[0127] The point coding set is P = {a, b, c}. Construct a mapping function f, such as f(a) = 1, f(b) = 2, f(c) = 3.

[0128] Step1.2:

[0129] For M1, its point position coding set is P1 = {a, b, c}, and the mapped feature quantity set is F1 = {1, 2, 3}. Calculate h(F1).

[0130] Step1.3:

[0131] Calculate the feature quantity sets and their hash values of M2, M3, and M4 in the above manner, and store them in H1.

[0132] Step1.4:

[0133] Construct the full feature quantity set A = {1, 2, 3}.

[0134] Step1.5:

[0135] For the element 1 ∈ A, find the fault model formulas containing this element, such as M1, M3, and M4. Calculate the hash values of their feature quantity sets and store them in H2(1). Finally, obtain the full hash value set B.

[0136] Step2:

[0137] Quickly filter the fault model formulas for the collected point positions;

[0138] Step2.1:

[0139] Suppose the collected point position coding set is P' = {a, b}. After mapping and sorting, the set C = {1, 2} is obtained.

[0140] Step2.2:

[0141] Calculate the intersection D = {1, 2}.

[0142] Step2.3:

[0143] Calculate the difference set E = {3}..

[0144] Step2.4:

[0145] According to E, search for H2(3) to obtain the hash value set F to be filtered.

[0146] Step 2.5:

[0147] Calculate the set H = B - F.

[0148] Step 3:

[0149] Extract the filtered data to implement the calculation of the effective fault model.

[0150] Step 3.1:

[0151] Traverse the set H. Let H = {h(F3, F4)}. Find the set of model formulas through H1(h(F3, F4)). In this example, it is {M3, M4}.

[0152] Step 3.2:

[0153] Substitute the collected data into M3 and M4 for calculation. Assume the collected data is D = {d1 = 1, d2 = 2}. Substituting into M3 gives 4×1 - 2 = 2, and substituting into M2 gives 1 + 5×2 = 11. Judge the fault status according to the calculation results.

[0154] By executing the above steps, the whole process from a complete monitoring point collection to matching the fault model formula is completed. For the matching of other collection points and fault models, steps S200 to S300 can be referred to.

[0155] Through the above steps, the present invention realizes the rapid matching calculation of the fault model in the field of intelligent operation and maintenance, can ensure the efficient matching and status judgment of the fault model during the point collection, avoids the matching and calculation of a large number of unrelated fault models, effectively improves the system fault diagnosis efficiency, and reduces the operation risk caused by untimely fault handling.

[0156] By accurately extracting the characteristic quantities of the fault model formula, optimizing the data organization form by using the personnel coding mapping sorting technology, establishing a multi-layer hash index storage method, realizing the rapid positioning and retrieval of data, and through effective filtering rules, a large number of unnecessary fault model matching calculations are successfully avoided, providing an efficient and rapid matching method for processing a large number of fault models, and significantly improving the overall efficiency of fault diagnosis and processing of the intelligent operation and maintenance system.

[0157] On the other hand, the present invention also discloses a computer-readable storage medium storing a computer program, which when executed by a processor, causes the processor to execute the steps of the above method.

[0158] In another aspect, the present invention also discloses a computer device, including a memory and a processor. The memory stores a computer program. When the computer program is executed by the processor, the processor is caused to execute the steps of the above method.

[0159] In yet another embodiment provided by the present application, there is also provided a computer program product containing instructions. When it runs on a computer, the computer is caused to execute any one of the above-described methods for quickly matching a fault model based on feature quantity extraction and multi-layer hash storage.

[0160] It can be understood that the system, device, and storage medium provided by the embodiments of the present invention correspond to the method provided by the embodiments of the present invention. For the explanations, examples, and beneficial effects of the relevant content, reference can be made to the corresponding parts in the above method.

[0161] In the above embodiments, it can be implemented in whole or in part by software, hardware, firmware, or any combination thereof. When implemented using software, it can be implemented in whole or in part in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable devices. The computer instructions can be stored in a computer-readable storage medium, or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center in a wired manner (such as coaxial cable, optical fiber, digital subscriber line (DSL)) or wirelessly (such as infrared, wireless, microwave, etc.). The computer-readable storage medium can be any available medium that can be accessed by a computer, or a data storage device such as a server or data center that includes one or more integrated available media. The available medium can be a magnetic medium (such as a floppy disk, hard disk, magnetic tape), an optical medium (such as a DVD), or a semiconductor medium (such as a solid state disk (SSD)).

[0162] It should be noted that in this document, relational terms such as first and second are only used to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any actual relationship or order between these entities or operations. Moreover, the terms "comprising", "including" or any other variant thereof are intended to cover non-exclusive inclusion, such that a process, method, article or device comprising a series of elements not only includes those elements but also includes other elements not expressly listed, or elements inherent to such process, method, article or device. Without further limitation, an element defined by the phrase "comprising an..." does not exclude the presence of additional identical elements in the process, method, article or device comprising said element.

[0163] Each embodiment in this specification is described in a related manner. For the parts that are the same or similar among the embodiments, reference can be made to each other. Each embodiment focuses on the differences from other embodiments. In particular, for system embodiments, since they are basically similar to method embodiments, the description is relatively simple, and reference can be made to the relevant parts of the method embodiments for the related parts.

[0164] The above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although the present invention has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that they can still modify the technical solutions described in the foregoing embodiments, or perform equivalent replacements for some of the technical features. However, such modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A fast matching method for fault models based on feature quantity extraction and multi-layer hash storage, characterized in that It includes the following steps: S100. Encode the point positions of the fault model formula and extract feature quantities; S200. Quickly filter the feature quantities of the collected point positions through the fault model formula; S300. Extract the filtered data to implement effective fault model calculation.

2. The fast matching method for a fault model based on feature quantity extraction and multi-layer hash storage according to claim 1, characterized in that The method for encoding the point positions of the fault model formula and extracting feature quantities in step S100 includes: Let the set of fault model formulas be \(M = \{M_1, M_2, M_3, \cdots, M\}\), where \(M\) n represents the \(i\)-th fault model formula. For each fault model formula \(M\) i , its set of point position codes is \(P\) i =\(\{p\) i , p\) i1 , p\) i2 , p\) i3 , \cdots p\) im \}\), where \(p_i\) j represents the \(j\)-th point position code in the \(i\)-th fault model formula; S110. Construct a dictionary to map the encoding of each point to a unique integer; Let the set of point codes be p = {p1, p2, p3,... p k}, construct a mapping function f: P -> Z, and map each point code p i to a unique integer f(p j ) = z j , where Z represents the set of integers; S120. Extract the point position encodings in each fault model formula, obtain the set of feature quantities after mapping for each fault model, and establish a hash storage for the hash value of the result and the fault model formula; For each fault model formula M i , the set of mapped characteristic quantities is as follows: Fi = {f(p i1 ), f(p i2 ), f(p i3 ),..., f(p im )} Calculate the hash value h(F i ), where the hash function is h: 2 Z -> H, H represents the set of hash values. Using h(F i ) as the key value, establish a hash storage H1 with the set of fault model formulas containing the same hash value as the key value, satisfying: H1(h(F i )) = {M j |h(F j ) = h(F i ), j = 1, 2,... n}; S130. Merge and deduplicate the sets of feature quantities in all fault models to obtain the full set of feature quantities A of the fault model formula; Set the full set of feature quantities The deduplication operation is expressed as A = {a|a ∈ Fi, i = 1, 2,... n}, where a is unique; S140. Establish a hash storage for the set of hash values of the sets of feature quantities of all fault model formulas containing each element of the full set of feature quantities A; For each element a ∈ A, find all the sets of fault model formulas containing this element M a = {M j | a ∈ F j , j = 1, 2,... n}, Calculate the hash value set H of its feature quantity set a ={h(F j )|M j ∈M a}, use a as the key and H a as the key value to establish a hash storage H2, satisfying: H2(a)=H a ; S150. Merge and deduplicate the hash values of the sets of feature quantities in all fault models to obtain the full set of hash values B of the fault model formula; Let \(B = \{h(F i )|i = 1,2,...n\}\), and the duplicate removal operation is expressed as \(B=\{b|b = h(F i ),i = 1,2,...,n\}\), where \(b\) is unique.

3. The fast matching method for fault models based on feature quantity extraction and multi-layer hash storage according to claim 2, wherein The method for establishing hash storage in step S120 includes: S121. Extract the set of feature quantities after mapping for each fault model and calculate the hash value of the set of feature quantities; For the fault model formula M i , obtain the feature quantity set F according to the method i and calculate h(F i ), where the hash function h satisfies: where hash represents the hash calculation function, and SHA-256 is selected; S122. Establish a hash storage with the hash value of the set of feature quantities as the key value and the set of model formulas with the same hash value for all sets of feature quantities as the key value; H1 defined in the above step S120 satisfies: H1(h(F i )) = {M j | h(F j ) = h(F i ), j = 1, 2,... n}.

4. The rapid matching method for fault models based on feature quantity extraction and multi-layer hash storage according to claim 3, characterized in that, The method for establishing hash storage in step S140 includes: For each element a ∈ A, find all the sets of fault model formulas containing this element M a = {M j | a ∈ F j , j = 1, 2,... n}, Calculate the hash value set H of its feature quantity set a ={h(F j )|M j ∈M a}, with a as the key and H a as the key value to establish a hash storage H2, satisfying: H2(a)=H a ; S141: Extract each element of the full set of feature quantities; For each element a ∈ A, where A is the full set of feature quantities; S142: Find all the fault models containing this feature quantity; Find the set M a ={M j |a ∈ F j , j = 1, 2,... n}; S143: Calculate the hash value of the set of feature quantities of all the matching fault models; For each M j ∈ M a , calculate the hash value h(F j ) of its feature quantity set F j ); S144: Establish a hash storage with each element of the full set of feature quantities as the key value and the set of hash values of the sets of feature quantities of all the matching fault model formulas as the key value.

5. The fast matching method for a fault model based on feature quantity extraction and multi-layer hash storage according to claim 1, wherein The method for quickly filtering the feature quantities of the collected point positions is as follows: S210. Map the collected point position encodings to unique integers according to the constructed dictionary and sort them in ascending order to obtain set C; Let the set of point codes collected be P' = {p'1, p'2, p'3,... p' s}, and the mapped set be C = {f(p'1), f(p'2), f(p'3),... f(p' s )}, and perform a sorting operation on C, satisfying: C = sort({f(p' j ) | p' j ∈ P'}) S220. Obtain the intersection D of the set of feature quantities A in step S130 and set C; Set D = A ∩ C; S230. Obtain the set of feature quantities E after removing the intersection D from the set of feature quantities A; Set E = A - D; S240. Find the set of all hash values F to be filtered according to the set of feature quantities E by looking up the hash table established in step S144; Among them, the set F is: S250. Obtain set H after removing the set of hash values F from the set of hash values B, that is, obtain the set of hash values of the sets of feature quantities of all the model formulas that need to calculate the point positions; Set H = B - F.

6. The fast matching method for a fault model based on feature quantity extraction and multi-layer hash storage according to claim 1, characterized in that The method for extracting and implementing an effective fault model in step S300 is as follows: S310. Traverse set H; S320. The hash table fast query model formula established through step S122; Obtain the model formula set M by querying with H1(h) h = H1(h); S330. Substitute data into the model formula for calculation; For each M i ∈ M h , substitute the collected data into formula M i for calculation. Assume the collected data is D = {d1, d2, d3,... d t}. Substitute the data into the fault model formula M i for fault diagnosis and calculation to determine the fault status.

Images