Abnormality detection system and abnormality detection method

Abstract

To provide an abnormality detection device and an abnormality detection method that can quickly detect an abnormality in a control system, and can specify and visualize a portion of the abnormality, which employ a modeling technique of a multilevel flow model, even when the control system is complex and its time-sequence changes are large compared to, for example, a plant.SOLUTION: An abnormality detection system 1 detects an abnormality in a control system, which is connected in hardware to a predetermined motor system (an example of the control system) that is configured by including an electric circuit system (an example of a circuit system) and actually operates after modeling the control system on software. The abnormality detection system 1 is configured by including a model setting unit 16A, which is provided with a plurality of functional elements F each of which corresponds to a function of the control system according to a modeling technique of a multilevel flow model.SELECTED DRAWING: Figure 5

Description

[Technical field]

[0001] The present invention relates to an anomaly detection system and an anomaly detection method that are connected in hardware to a specific control system that is configured including a circuit system and actually operates, model the control system in software, and detect an anomaly in the control system. [Background technology]

[0002] As a conventional technique for detecting an abnormality in a given control system that is actually operating, a system is known that includes a state quantity measuring device that measures a state quantity of the control system, a measured quantity determining device, a diagnostic device, and a display device (see, for example, Patent Document 1). The control system that is the target of the device described in Patent Document 1 is a plant.

[0003] The measured quantity determination device described in Patent Document 1 determines a state quantity corresponding to a plant model to be diagnosed for abnormality based on a measurement signal from a state quantity measuring device, and determines whether the plant is normal or abnormal based on a reference value and a threshold value with reference to a plant operation schedule from a plant control device. The diagnostic device described in Patent Document 1 inputs the determination result of the measured quantity determination device and identifies abnormal elements in the plant model according to a predetermined diagnostic method. The display device described in Patent Document 1 displays information indicating the abnormality diagnosis result and inference process of the diagnostic device.

[0004] As another prior art, a system is also known in which the target control system is a plant, similar to Patent Document 1, and which includes a state quantity measuring device that measures the state quantity of the control system, a measured quantity determination device, a diagnostic device, and a display device (see, for example, Patent Document 2).

[0005] The measured quantity determination device described in Patent Document 2 determines whether a measured state quantity related to a plant function hierarchical model that is the target of an abnormality diagnosis is normal / abnormal based on a measurement signal from a state quantity measuring device. The measured quantity determination device described in Patent Document 2 also changes a threshold value for determining whether the measured state quantity is normal / abnormal and determines the priority of each goal in the plant function hierarchical model in accordance with a plant operation schedule from a plant control device. The diagnosis device described in Patent Document 2 determines an abnormality propagation network on the plant function hierarchical model based on the determination result of the measured quantity determination device and performs abnormality diagnosis. The display device described in Patent Document 2 displays the abnormality diagnosis result and inference process of the diagnosis device. [Prior art documents] [Patent documents]

[0006] [Patent Document 1] Japanese Patent Application Publication No. 9-190219 [Patent Document 2] Japanese Patent Application Publication No. 10-20932 Summary of the Invention [Problem to be solved by the invention]

[0007] Incidentally, multi-level flow models (MLFMs) are known as a modeling technique for modeling a control system and adaptively controlling it.

[0008] MLFM is used in process control or control system design, where real physical elements are represented as elements in the model, allowing the model to more directly correspond to the real world. Furthermore, MLFM is capable of expressing system dynamics. In other words, MLFM can appropriately express feedback loops and time-series changes within a control system, making it possible to accurately analyze the dynamics of the control system, such as stability or oscillation, growth or decay of the control system. Furthermore, MLFM can appropriately divide and integrate the complexity of the control system, making it easier for operators to understand and analyze the system, even when dealing with large-scale and complex control systems.

[0009] Thus, although MLFM is a modeling technique with various advantages, conventional systems such as those described in Patent Document 1 and Patent Document 2 above do not support MLFM as a modeling technique. For example, the device in Patent Document 1 is based on a neural network that represents the flow of mass energy. It can be said that there is room for improvement in terms of application to MLFM. Also, as described above, even if the control system is complex and the changes in time series are large compared to, for example, a plant, it is possible to respond adaptively and quickly by adopting the model technique of MLFM in the anomaly detection system.

[0010] The present invention has been made in consideration of the above-mentioned circumstances, and has as its object to provide an anomaly detection device and an anomaly detection method that employ a multilevel flow model modeling technique and are capable of quickly detecting an anomaly in a control system and identifying and visualizing the location of the anomaly, even when the control system is complex and the changes over time are large compared to, for example, a plant. [Means for solving the problem]

[0011] The above-mentioned object of the present invention can be achieved by the following configuration. [1] An anomaly detection system that is connected in a hardware manner to a predetermined control system that is configured including a circuit system and actually operates, models the control system in software, and detects anomalies in the control system, a model setting unit which provides a plurality of functional elements corresponding to each of the functions of the control system in accordance with a multilevel flow model technique, and sets one main group represented by the plurality of functional elements connected in series with each other and through which information flows serially from the upper end to the lower end, and one or more subgroups represented by the plurality of functional elements connected directly with each other and through which information flows serially from the upper end to the lower end, each of which is connected to any one of the plurality of functional elements of the main group at its most distal functional element; an actual state acquisition unit that acquires actual state quantities of the functional elements in the main group and each of the subgroups and expresses the actual state quantities in a matrix format; a virtual state calculation unit that calculates, in a matrix format, a number of virtual state quantities in the main group and each of the plurality of subgroups, the same number as the number of the functional elements belonging to the main group or each of the plurality of subgroups; a state comparison unit that compares each of the plurality of virtual state quantities calculated by the virtual state calculation unit with an acquired result by the real state acquisition unit; and an abnormality identification unit that identifies the functional element in which an abnormality has occurred based on a comparison result of the state comparison unit. Anomaly detection system. [2] The plurality of subgroups are set up to an L-th subgroup, and the plurality of functional elements are set up in a j-th subgroup (j is a natural number; j=1, . . . , L) from the top end along the information flow to an M-th subgroup (M is a natural number), The actual state acquisition unit acquires the actual state quantity S of the k-th (k is a natural number; k=1, . . . , M)-th functional element in the j-th subgroup. jk(indicating the state quantity of the k-th functional element in the j-th subgroup) in vector form, and a sub-whole state quantity P indicating the actual state quantity of the j-th subgroup as a whole. j is expressed in matrix form according to Equation 1 below, The virtual state calculation unit calculates the plurality of virtual state quantities Q jk is calculated according to the following formula 2, The state comparison unit compares the plurality of virtual state quantities Q jk For each of the subgroups, the sub-total state quantity P j to determine whether they match, [1] The anomaly detection system described in. [Formula 1] P j =(S j1 S j2 ··· S j(M-1) S jM ) where j is a natural number; j=1, ,L [Formula 2] Q jk =(S j1 ··· S j(k-1) S jk T1 1 ·S jk T1 M-k ·S jk ) where j and k are both natural numbers; j=1,···,L; k=1,···,M T j is a virtual abnormal transfer function set between the functional elements that are adjacently arranged in the j-th subgroup. [3] an abnormality detection unit that detects an abnormality in a function of the control system based on a result of the identification by the abnormality identification unit, The virtual state quantity Q jkis set to correspond to the k-th functional element in the j-th subgroup, When there is only one match in the j-th subgroup as a result of the comparison by the state comparison unit, the anomaly identification unit identifies the matched virtual state quantity Q jk Based on the virtual state quantity Q jk or, if there are a plurality of coincident virtual state quantities Q jk Based on this, the virtual state quantity Q jk identifying the functional element arranged at the top among the corresponding functional elements; The abnormality detection unit detects that an abnormality has occurred in a function of the control system corresponding to the functional element identified by the abnormality identification unit. [2] The anomaly detection system described in. [4] In the main group, when the plurality of functional elements are counted from the top to the end of the information flow, Nth (N is a natural number), The actual state acquisition unit acquires the state quantity U of the i-th (i is a natural number; i=1, . . . , N) functional element in the main group. i (indicating the state quantity of the i-th functional element in the main group) in vector form, and expressing a main overall state quantity V indicating the state quantity of the entire main group in matrix form according to Equation 3 described later; The virtual state calculation unit calculates the plurality of virtual state quantities W i is calculated according to the following formula 4, Furthermore, the virtual state calculation unit calculates the virtual state quantities W iFor each of the matrix elements arranged in the horizontal direction in Equation 4 described later, the actual state quantity S of the functional element arranged at the end of the y-th subgroup is determined to be the matrix element that is the next matrix element after the x-th matrix element and is calculated by matrix multiplying a virtual anomalous transfer function T described later. y Then, we multiply each of these by a matrix to obtain multiple virtual state quantities W i 'Calculate (i=1, ,N), The state comparison unit compares the plurality of virtual state quantities W i ' to compare the main overall state quantity V with each of the main overall state quantities V and V, and determine whether they match each other; [1] The anomaly detection system described in. [Formula 3] V = (U1U2 U N-1 U N ) [Formula 4] W i = (U1 U i-1 U i T 1 ·U i T N-i ·U i ) where i is a natural number; i=1, ,N T is a virtual abnormal transfer function set between the functional elements adjacently arranged in the main group. [5] an abnormality detection unit that detects an abnormality in a function of the control system based on a result of the identification by the abnormality identification unit, The virtual state quantity W i ' is set to correspond to the i-th functional element in the main group, When only one virtual state quantity W matches as a result of the comparison by the state comparison unit, the anomaly specification unit i Based on the virtual state quantity W i', or, if there are a plurality of coincidents, the coincident plurality of virtual state quantities W i Based on this, the virtual state quantity W i ', identify the functional element that is arranged at the top among the functional elements corresponding to The abnormality detection unit detects that an abnormality has occurred in a function of the control system corresponding to the functional element identified by the abnormality identification unit. The anomaly detection system described in [4]. [6] The virtual anomalous transfer function is an identity matrix. The anomaly detection system according to claim 2 or 4. [7] An anomaly detection method for detecting an anomaly in a control system that is actually operating and includes a circuit system, by connecting the control system in a hardware manner using a predetermined device and / or program, and modeling the control system in software, comprising: a model setting step of providing a plurality of functional elements corresponding to each of the functions of the control system in accordance with a multilevel flow model technique, and setting one main group represented by the plurality of functional elements connected in series with each other and through which information flows serially from the upper end to the lower end, and one or more subgroups represented by the plurality of functional elements connected directly with each other and through which information flows serially from the upper end to the lower end, each of which is connected to any one of the plurality of functional elements of the main group at its most distal functional element; an actual state acquisition step of acquiring actual state quantities of the respective functional elements in the main group and the respective subgroups and expressing the actual state quantities in a matrix format; a virtual state calculation step of calculating, in a matrix format, a number of virtual state quantities equal to the number of the functional elements belonging to the main group or the subgroups, in each of the main group and the subgroups; a state comparison step of comparing each of the plurality of virtual state quantities calculated in the virtual state calculation step with an acquired result in the actual state acquisition step; and an abnormality identification step of identifying the functional element in which an abnormality has occurred based on a comparison result in the state comparison step. Anomaly detection methods. Effect of the Invention

[0012] According to the present invention, a modeling technique of a multilevel flow model is adopted, and even if the control system is complex and the changes over time are large compared to, for example, a plant, it is possible to quickly detect abnormalities in the control system and to identify and visualize the location of the abnormality.

[0013] The present invention has been briefly described above. Furthermore, the details of the present invention will be more clearly understood by reading the following description of the preferred embodiment of the present invention (hereinafter referred to as "embodiment") with reference to the accompanying drawings. [Brief description of the drawings]

[0014] [Figure 1] FIG. 1 is a schematic configuration diagram illustrating an example of a network configuration related to an anomaly detection system according to a first embodiment of the present invention. [Diagram 2] FIG. 2 is a schematic diagram illustrating an example of the configuration of a vehicle steering device that is the target of control by the abnormality detection system shown in FIG. [Diagram 3] FIG. 3 is a circuit diagram illustrating an example of a schematic configuration of an electric circuit of the vehicle steering device shown in FIG. [Figure 4] FIG. 2 is a schematic diagram illustrating an example of a hardware configuration of the server device illustrated in FIG. 1. [Diagram 5] FIG. 2 is a block diagram illustrating an example of functions realized by the server device and the information terminal device shown in FIG. 1. [Figure 6] FIG. 3 is a functional block diagram illustrating an example of a control model that represents the control target of FIG. 2 by a model setting unit. [Figure 7] FIG. 6 is a flow diagram illustrating an example of an operation flow executed by the server device illustrated in FIG. 5. DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS

[0015] DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS One or more embodiments specifically disclosing an anomaly detection system and an anomaly detection method according to the present invention will be described in detail below with appropriate reference to the accompanying drawings.

[0016] However, more detailed explanations than necessary may be omitted. For example, detailed explanations of already well-known matters or duplicate explanations of substantially the same configuration may be omitted. This is to avoid the following explanation becoming unnecessarily redundant and to facilitate understanding by those skilled in the art. In addition, each of the attached drawings should be viewed according to the direction of the reference numerals.

[0017] Furthermore, the accompanying drawings and the following description are provided to enable those skilled in the art to fully understand the present disclosure, and are not intended to limit the subject matter described in the claims.

[0018] <Terminology explanation> The terms "comprising" or "characterized by" and "characterized by," which are synonymous with "having," "comprising," "including," and "containing," are to be interpreted in an inclusive or open-ended sense and do not exclude additional, unrecited elements or method steps. "Comprising" is a term of art used in claim language that means that the named claim element is required, but that other claim elements may be added to further form a structure within the scope of the claim.

[0019] Also, when the phrase "consisting of" is used herein, it excludes any element, step, or ingredient not specified in the claim. When the phrase "consisting of (or variations thereof)" appears in a section of the body of a claim rather than immediately following the preamble, it limits only the elements set forth in that section and does not exclude other elements from the claim as a whole. As used herein, the phrase "consisting essentially of" limits the scope of a claim to those elements or method steps specified in addition to those that do not materially affect the main and novel feature(s) of the claimed subject matter.

[0020] With respect to the terms "comprising," "consisting of," and "consisting essentially of," when one of these three terms is used herein, the presently disclosed and claimed subject matter may also include the use of either of the other two terms. Thus, in some embodiments not expressly recited otherwise, any instance of "comprising" may be replaced by "consisting of" or "consisting essentially of."

[0021] The term "process" or "step" may be used explicitly or implicitly in connection with a process or method feature, but no limitation is imposed on the order or sequence between such explicit processes or steps or between implicit processes or steps, unless the order or sequence is stated.

[0022] In addition, the terms "unit" and "device" used in this specification are not limited to a physical configuration that is mechanically realized by hardware, but also include configurations whose functions are realized by software such as programs. Also, the functions of one configuration may be realized by two or more physical configurations, or the functions of two or more configurations may be realized by, for example, one physical configuration (for example, one independent device).

[0023] Furthermore, the term "user" used in the following description of the embodiments is not limited to referring to an inspector who owns an information terminal device (e.g., a dedicated terminal or a computer such as a smartphone, tablet, or laptop) and appropriately inspects a controlled object (e.g., a vehicle, a vehicle steering device, etc.), but is broadly interpreted to include the driver of a vehicle or its owner, as described below.

[0024] <Multi-level flow model> Multi-Level Flow Models (MLFM) is a formal modeling method for systems used in process control and control system design. The following describes its characteristics and their explanations in bullet points.

[0025] [·Hierarchical structure] MLFM has a hierarchical structure. This structure allows the creation of a more concise and understandable model by dividing the system from the top to the bottom. Each level can be divided into more detailed levels, allowing the complexity of the system to be managed effectively. For example, the top level of MLFM can represent the entire system, while the intermediate levels can represent partial components of the system. The lowest level can represent fine details of the components.

[0026] [·Physically meaningful expression] MLFM uses representations that have physical meaning. The main purpose of this model is that the actual physical elements in a process or control system are represented as elements in the model, allowing the model to correspond more directly to the real world. This feature is one of the things that distinguishes MLFM from other formal methods.

[0027] [·Multiple levels of control] MLFM allows control at multiple levels. The model represents the state at each level and control can be exercised at each level. For example, MLFM can exercise partial control and, if partial control is successful, it can proceed to the next stage toward total control of that subsystem.

[0028] [·Versatility] MLFM is a versatile model and can be used for different purposes. It can be used for design and control, troubleshooting, and functional analysis of systems. MLFM is a very flexible model and can be used for many purposes.

[0029] [·Intuitive expression] MLFM uses an intuitive representation method, which makes it easier to understand and interpret the model. In addition, MLFM can use simple and clear representations, so even complex models can be intuitively understood. In addition, MLFM uses representations that have physical meaning, so in that sense, it is easy to understand intuitively. MLFM clearly represents the relationships between components, so it is possible to more clearly explain the roles of elements in the model.

[0030] [High readability (visibility)] MLFM has the characteristic of being highly readable. MLFM uses hierarchical and physically meaningful expressions, which makes it easy to understand the model. In many cases, it can be visually expressed using diagrams and charts, making it intuitive to understand. Therefore, MLFM can provide high readability for various purposes of system analysis, such as design and troubleshooting.

[0031] First Embodiment A first embodiment of an anomaly detection system 1 and an anomaly detection method according to the present invention will be described with reference to FIGS. 1 to 7. FIG.

[0032] [Network configuration of the anomaly detection system] The network configuration of an anomaly detection system 1 of this embodiment will be described with reference to FIG. FIG. 1 is a schematic diagram illustrating an example of a network configuration related to an anomaly detection system 1 of the present embodiment.

[0033] As shown in FIG. 1, the anomaly detection system 1 of this embodiment includes a communication network N, a plurality of server devices 10, and a plurality of information terminal devices 20 mounted on a vehicle C. In this embodiment, a vehicle steering device 30 (see below) is mounted on the vehicle C, and the vehicle steering device 30 is set as a control target (anomaly detection target) in the anomaly detection system 1 of this embodiment.

[0034] The server devices 10 are arranged, for example, in a predetermined facility. The information terminal device 20 is mounted, for example, in a vehicle C and connected in terms of hardware so as to be operable by each of the users H.

[0035] With such a configuration, the server device 10 detects the overall condition of the vehicle C (in this embodiment, the vehicle steering device 30) from each information terminal device 20, and each user H can grasp whether or not an abnormality has occurred in the vehicle C through the information terminal device 20. In addition, if the information terminal device is a small terminal such as a smartphone, the information terminal device may be configured to be connected to the vehicle C in a hardware manner when necessary, and to be portable or movable (relocatable) at other times.

[0036] The communication network N is connected to each of the multiple server devices 10 and the multiple information terminal devices 20, and is provided so that various information or signals can be exchanged between them. The communication network N includes at least wired or wireless lines, and is configured to include Internet lines as a part or all of them. As a result, the communication network N is configured so that communication can be performed between the server devices 10 and the information terminal devices 20, and between the information terminal devices 20. In other words, the multiple server devices 10 are configured so that they can be accessed by the information terminal devices 20 of the multiple users H through the communication network N.

[0037] In addition, the communication network N may also include a local area network (LAN), a wide area network (WAN), a mobile telephone network (MTN), and other types of networks that are interconnected and used to facilitate communication between the server device 10 and the information terminal device 20, as appropriate.

[0038] As described below, the information terminal device 20 is configured to have at least an image output unit 24 (e.g., a monitor device or a display device), an operation input unit 23 (e.g., a touch panel, a keyboard, or a mouse), and an audio output unit (e.g., a speaker: not shown). Furthermore, the information terminal device 20 is configured to be able to communicate information with the server device 10 via the communication network N, as described above. Therefore, the information terminal device 20 is able to output information relating to an abnormality detected by the server device 10 to the image output unit 24 and the audio output unit, and present (notify) the user H of the detection status.

[0039] In this embodiment, the information terminal device 20 is exemplified by a minicomputer composed of one or more semiconductors, but is not limited thereto. For example, the information terminal device 20 can be adopted in various other devices, such as a notebook computer, a smartphone, a general-purpose personal computer device, a mobile phone, a tablet, a laptop computer, a netbook, a personal digital assistant (PDA), a game device, a media player, or an electronic book. The form of the information terminal device 20 is not limited as long as it can be connected (accessed) to the communication network N.

[0040] In this embodiment, the main functions of the anomaly detection system 1 according to the present invention are implemented in the server device 10. The server device 10 communicates with a plurality of information terminal devices 20 via a communication network N to detect the occurrence of an anomaly in each of the vehicles C, and notifies the detection result to, for example, the driver of the vehicle C or an on-site operator via the information terminal device 20.

[0041] In this embodiment, most of the functions of the anomaly detection system 1 are implemented in the server device 10, but this is not limited to the above. For example, some or all of the functions of the server device 10 may be implemented in the information terminal devices 20, and each of the information terminal devices 20 may be configured to autonomously realize most of the functions of the system.

[0042] [Outline of vehicle steering system configuration] The configuration of a vehicle steering device 30 according to this embodiment will be described with reference to FIGS. FIG. 2 is a schematic diagram illustrating an example of the configuration of the vehicle steering device 30 that is the control target of the abnormality detection system 1 shown in FIG. FIG. 3 is a circuit diagram illustrating an example of a schematic configuration of an electric circuit of the vehicle steering device 30 shown in FIG.

[0043] The vehicle steering device 30 includes a steering wheel (not shown) that is steered by a driver (which may be the above-mentioned user H).

[0044] As shown in FIG. 2, the steering shaft 31 of the steering wheel is connected to steered wheels 37 through a reduction gear 32 (worm gear) that constitutes a reduction mechanism and is built into a motor 38 described later, a universal joint 33, a pinion rack mechanism 34, a tie rod 35, and further through a hub unit 36.

[0045] The steering shaft 31 is configured by connecting an input shaft on the steering wheel side and an output shaft on the pinion rack mechanism 34 side via a torsion bar (not shown).

[0046] The pinion rack mechanism 34 has a pinion 34A and a rack 34B that meshes with the pinion 34A. The pinion 34A is connected to a pinion shaft (not shown) to which steering force is transmitted from the universal joint 33. The rotational motion transmitted to the pinion 34A is converted into linear motion in the vehicle width direction by the rack 34B.

[0047] The vehicle steering device 30 further includes a motor 38 (an example of a circuit system), a rotation sensor 39 (an example of a circuit system), an ECU 40 (Electronic Control Unit, an example of a circuit system), and a battery 41 as a power source.

[0048] The motor 38 is connected to the steering shaft 31 via the reduction gear 32. The motor 38 transmits the steering force of the vehicle C to the steering wheel. The motor 38 is, for example, a three-phase AC motor, and has a permanent magnet field or a wound field. The motor 38 is driven to rotate by supplying three-phase AC currents, the phases of which differ by 120°, to coils (not shown) of u-phase, v-phase, and w-phase.

[0049] A rotation sensor 39 including a resolver (not shown, an example of a circuit system) and a rotary encoder (not shown, an example of a circuit system) is disposed on the shaft of a rotor (not shown) of the motor 38. The rotation sensor 39 includes a plurality of circuits as such an electric circuit system, and detects the rotation angle or rotation speed of the motor 38.

[0050] The ECU 40 is a controller that controls the vehicle steering device 30, and is supplied with power from a battery 41 (an example of a circuit system) to the ECU 40. The power is also supplied to the motor 38 via the ECU 40. The rotation angle of the motor 38 detected by a rotation sensor 39 is output to the ECU 40.

[0051] That is, the ECU 40 controls the power supplied to the motor 38 based on the detection information of the rotation sensor 39, and as a result, can issue a drive command for the motor 38. Then, the motor 38 operates the reduction gear 32 based on the drive command from the ECU 40, thereby achieving, for example, assist control for the steering wheel. In this way, the ECU 40 performs drive control of the motor control system in the vehicle steering device 30.

[0052] That is, in this embodiment, a motor control system is exemplified as the "predetermined control system that includes a circuit system and actually operates" as set forth in the claims of the present invention. However, as described above, in this embodiment, a motor control system is shown as an example of the present invention, but is not limited thereto. Its application as a controlled object is wide-ranging, such as a plant system such as an oil plant, and a control system of a smartphone that is entirely composed of electric circuits or electronic circuits.

[0053] The ECU 40 is connected across other electric and electronic circuits and / or devices inside the vehicle C via a CAN (Controller Area Network) and receives, for example, steering target value information from a central management unit of the vehicle C, and drives and controls the motor 38 in accordance with the target value information.

[0054] Furthermore, the ECU 40 may be configured to have, for example, a computer including a processor and peripheral components (electrical and electronic circuits) such as the storage device 12.

[0055] The processor of the ECU 40 may be, for example, a central processing unit (CPU) or a micro-processing unit (MPU). The storage device of the ECU 40 may include any one of a semiconductor storage device, a magnetic storage device, and an optical storage device.

[0056] The storage device of the ECU 40 may include memories such as a register, a cache memory, a read only memory (ROM) used as a main storage device, and a random access memory (RAM), etc. The functions of the ECU 40 are realized, for example, by a processor of the ECU 40 executing a computer program stored in the storage device.

[0057] The ECU 40 may be configured with dedicated electric and electronic hardware for executing each information processing. For example, the ECU 40 may have a functional logic circuit set in a general-purpose semiconductor integrated circuit. For example, the ECU 40 may be configured to include a programmable logic device (PLD) such as a field programmable gate array (FPGA).

[0058] In this embodiment, the ECU 40 is connected to the information terminal device 20, and control information (all or part of information including drive information and detection information) in the ECU 40 is transmitted to and received from the information terminal device 20. As described above, the information terminal device 20 transmits and receives the control information to and from the server device 10 via the communication network N.

[0059] In this way, the anomaly detection system 1 is connected in a hardware manner to a motor control system, which is exemplified as a control system in this embodiment. Then, the anomaly detection system 1 models the control system in software according to a multilevel flow model technique, as described below, and detects an anomaly in the control system.

[0060] [About the hardware configuration of the server device] The hardware configuration of the server device 10 of this embodiment will be described with reference to FIG. FIG. 4 is a schematic diagram illustrating an example of a hardware configuration of the server device 10 shown in FIG.

[0061] 4, each of the server devices 10 is configured by a so-called general-purpose server computer. The server device 10 includes a CPU 11 (Central Processing Unit, an example of a processing device), a storage device 12, and a communication device 15.

[0062] The CPU 11 of the server device 10 executes programs that realize various functions described later as the central processing unit thereof, whereby the CPU 11 of the server device 10 operates functionally as a control unit 16 of the entire computer (see below).

[0063] The storage device 12 of the server device 10 is not only used to store and hold programs, but is also used as a working memory for the CPU 11, and is configured to include various storage devices 12 such as a ROM (Read Only Memory) and a RAM (Random Access Memory). ROM is an example of a non-volatile semiconductor memory (storage device) that can hold programs and data even when the power is turned off. The ROM stores programs and data such as the BIOS, OS settings, and network settings that are executed when the computer starts up. RAM is an example of a volatile semiconductor memory (storage device) that temporarily holds programs and data.

[0064] The storage device 12 of the server device 10 may be configured to include a secondary storage device {e.g., a hard disk drive (HDD), a solid state drive (SSD)} or a tertiary storage device (e.g., an optical disk, an SD card), or may be configured to include other storage devices. The HDD is an example of a non-volatile storage device that stores programs and data. The stored programs and data include an OS, which is basic software for controlling the entire computer, and applications that provide various functions on the OS.

[0065] By executing such a program on the server device 10, the server device 10 can detect abnormalities in each vehicle steering device 30 through the information terminal device 20 by receiving the control information of the ECU 40 described above through the information terminal device 20.

[0066] The communication device 15 of the server device 10 is connected to each of the multiple information terminal devices 20 via the communication network N. The server device 10 exchanges various processes or communications required to detect an abnormality in the vehicle steering device 30 with each of the multiple information terminal devices 20 via the communication network N and the communication device 15 of the server device 10. As a specific example, the communication device 15 of the server device 10 transmits and receives information related to the control of the vehicle steering device 30 (the above-mentioned control information) transmitted by each of the multiple information terminal devices 20.

[0067] As described above, the communication method used by the communication device 15 of the server device 10 may be, for example, a wired communication method such as a WAN (Wide Area Network), a LAN (Local Area Network) or power line communication corresponding to the communication network N, or a wireless communication method such as Wi-Fi (Wireless Fidelity) or mobile communication for mobile phones. In addition, the communication method may be configured to include, as appropriate, Bluetooth (registered trademark) in addition to the above-mentioned communication methods.

[0068] The server device 10 further comprises an operation input device 13 and an image output device 14. The operation input device 13 of the server device 10 is, for example, a touch panel, operation keys, buttons, a keyboard, or a mouse used by an administrator or the like to input various signals. The image output device 14 of the server device 10 is a display device such as a liquid crystal or organic EL display that displays a screen.

[0069] In the server device 10, the devices are connected to each other via a bus BS. The operation input device 13 and the image output device 14 may be appropriately connected and used when necessary. The server device 10 is not limited to the general-purpose server computer described above, and may be configured by a general computer such as a tablet device, a smartphone, or a laptop computer, similar to the information terminal device 20.

[0070] [Functional configuration of server device and information terminal device] The functional configuration of the server device 10 and the information terminal device 20 of the user H will be described with reference to FIGS. FIG. 5 is a block diagram illustrating an example of functions realized by the server device 10 and the information terminal device 20 shown in FIG. FIG. 6 is a functional block diagram for explaining an example of a control model expressing the controlled object of FIG. 2 by the model setting unit 16A.

[0071] [··Server equipment] As described above, the server device 10 has a CPU 11 as a hardware configuration, and the CPU 11 realizes various functions by reading and executing various programs from the storage device 12. As a result, various processing programs related to abnormality detection of the vehicle steering device 30 in the abnormality detection system 1 are executed. Similarly, the information terminal device 20 has a central processing unit, and realizes various functions by reading and executing various programs from the storage device.

[0072] As shown in FIG. 5, in this embodiment, the server device 10 includes a storage unit 17, a communication unit 18, and a control unit 16 as functions realized by the operation of hardware such as the CPU 11.

[0073] The memory unit 17 of the server device 10 is a function realized by operation including the storage device 12 of the server device 10 as described above, and stores and holds information related to anomaly detection. The communication unit 18 of the server device 10 is a function realized by operation including the communication device 15 of the server device 10 as described above, and communicates with each of the information terminal devices 20. The control unit 16 of the server device 10 is a function realized by the CPU 11 reading programs and the like in the storage device 12 as described above, and performs processing related to anomaly detection as the anomaly detection system 1.

[0074] Then, the control unit 16 of the server device 10 executes abnormality detection based on the operation received by the information terminal device 20 from the user H and / or information regarding abnormality detection transmitted and received from the information terminal device 20 (including control information related to motor control).

[0075] The information terminal device 20 is connected in terms of hardware to the ECU 40 of the vehicle steering device 30 described above, that is, the server device 10 is connected in terms of hardware to a motor control system (an example of a predetermined control system that actually operates) controlled by the ECU 40 through the information terminal device 20. This enables the server device 10 to model the (controlled) motor control system (an example of a control system) in software, as described below, and then detect an abnormality in the motor control system.

[0076] The control unit 16 of the server device 10 includes a model setting unit 16A, a real state acquisition unit 16B, a virtual state calculation unit 16C, a state comparison unit 16D, an abnormality identification unit 16E, and an abnormality detection unit 16F.

[0077] In the following explanation, for convenience of explanation, the first functional element MF1, second functional element MF2, third functional element MF3, fourth functional element MF4, and fifth functional element MF5 of the main loop ML, the first functional element S1F1, second functional element S1F2, and third functional element S1F3 of the first sub-loop SL1, the first functional element S2F1, second functional element S2F2, and third functional element S2F3 of the second sub-loop SL2, and the first functional element S3F1, second functional element S3F2, and third functional element S3F3 of the third sub-loop SL3 may be collectively indicated by the symbol "F".

[0078] [...Model setting section] The model setting unit 16A of the server device 10 shown in FIG. 5 provides multiple functional elements F (also called "nodes" in information engineering, and this concept is also incorporated into the present invention) that individually correspond to each of the functions of the motor control system (an example of a control system) according to a multi-level flow model, and sets one main loop ML (an example of a main group, see below) and multiple (first to third) sub-loops SL1, SL2, SL3 (three in this embodiment, examples of sub-groups, see below).

[0079] The settings of the main loop ML and the first to third sub-loops SL1, SL2, and SL3 in this model setting unit 16A are configured to be realized by a system operator such as an engineer or a system administrator using the operation input device 13 of the server device 10 to perform modeling operations (model design) on specified software.

[0080] As shown in FIG. 6, in this embodiment, the above-mentioned motor control system has one main loop ML and first to third sub-loops SL1, SL2, and SL3, and is set by a model setting unit 16A. In this embodiment, three sub-loops SL1, SL2, and SL3 (one example of a plurality of sub-groups) are set, but the number is not limited to this and can be changed or set appropriately depending on the target control system.

[0081] The main loop ML represents a functional block that as a whole drives and controls the motor 38 in the motor control system according to the MLFM technique, and has the role of the main function in the motor control system described above. The main loop ML is represented by including a first functional element MF1, a second functional element MF2, a third functional element MF3, a fourth functional element MF4, and a fifth functional element MF5.

[0082] In the main loop ML, the first functional element MF1, the second functional element MF2, the third functional element MF3, the fourth functional element MF4, and the fifth functional element MF5 are connected in series with each other. The first functional element MF1 of the main loop ML is located at the top end, and the fifth functional element MF5 of the main loop ML is located at the bottom end. In the main loop ML, information flows serially from the first functional element MF1 of the main loop ML to the fifth functional element MF5 of the main loop ML.

[0083] According to the MLFM (Multi-Level Flow Model), a first functional element MF1 of the main loop ML is provided corresponding to the power supply function in the motor control system. A second functional element MF2 of the main loop ML is provided corresponding to the information transmission (signal transmission) function in the motor control system. A third functional element MF3 of the main loop ML is provided corresponding to the ECU 40 function in the motor control system. A fourth functional element MF4 of the main loop ML is provided corresponding to the transmission function in the motor control system. A fifth element of the main loop ML is provided corresponding to the motor 38 function in the motor control system.

[0084] The first sub-loop SL1, in its entirety, expresses the functional blocks of the CAN of the vehicle C according to the MLFM technique, and plays the role of a sub-function in the motor control system as a support function of the main loop ML. The first sub-loop SL1 is expressed by including a first functional element S1F1, a second functional element S1F2, and a third functional element S1F3.

[0085] In the first sub-loop SL1, as in the main loop ML, the first functional element S1F1, the second functional element S1F2, and the third functional element S1F3 are connected in series to each other. The first functional element S1F1 of the first sub-loop SL1 is located at the top end, and the third functional element S1F3 of the first sub-loop SL1 is located at the bottom end. Similarly, in the first sub-loop SL1, information flows serially from the first functional element S1F1 to the third functional element S1F3. And the third functional element S1F3 of the first sub-loop SL1, located at the end (bottom end), is connected to the third functional element MF3 of the main loop ML. In other words, the first sub-loop SL1 is connected to the main loop ML at (via) its third functional element S1F3.

[0086] Similarly, according to the MLFM, the first functional element S1F1 of the first sub-loop SL1 is provided corresponding to the function of information input (signal input) of the CAN in the motor control system. The second functional element S1F2 of the first sub-loop SL1 is provided corresponding to the function of information transmission in the motor control system. The third functional element S1F3 of the first sub-loop SL1 is provided corresponding to the function of information output (signal output) to the main loop ML in the motor control system.

[0087] The second sub-loop SL2 expresses the functional blocks of the battery 41 of the vehicle steering device 30 as a whole according to the MLFM technique, and has the role of a sub-function in the motor control system as a support function of the main loop ML. The second sub-loop SL2 is expressed by including a first functional element S2F1, a second functional element S2F2, and a third functional element S2F3.

[0088] In the second sub-loop SL2, similarly to the main loop ML, the first functional element S2F1, the second functional element S2F2, and the third functional element S2F3 are connected in series with each other. The first functional element S2F1 of the second sub-loop SL2 is located at the top end, and the third functional element S2F3 of the second sub-loop SL2 is located at the bottom end. Similarly, in the second sub-loop SL2, information flows serially from the first functional element S2F1 to the third functional element S2F3. And the third functional element S2F3 of the second sub-loop SL2, located at the end (bottom end), is connected to the first functional element MF1 of the main loop ML. In other words, the second sub-loop SL2 is connected to the main loop ML at its third functional element S2F3.

[0089] Similarly, according to the MLFM, a first functional element S2F1 of the second sub-loop SL2 is provided corresponding to the function of the battery 41 in the motor control system. A second functional element S2F2 of the second sub-loop SL2 is provided corresponding to the information transmission function in the motor control system (including the power transmission / transmission function). A third functional element S2F3 of the second sub-loop SL2 is provided corresponding to the information output (signal output) function in the motor control system (including the power output function).

[0090] The third sub-loop SL3 expresses the function block of the rotation sensor 39 of the vehicle steering device 30 as a whole according to the MLFM technique, and has the role of a sub-function in the motor control system as a support function of the main loop ML. The third sub-loop SL3 is expressed by including a first functional element S3F1, a second functional element S3F2, and a third functional element S3F3.

[0091] Similarly to the main loop ML, in the third sub-loop SL3, the first functional element S3F1, the second functional element S3F2, and the third functional element S3F3 are connected in series with each other. The first functional element S3F1 of the third sub-loop SL3 is located at the top end, and the third functional element S3F3 of the third sub-loop SL3 is located at the bottom end. Similarly, in the third sub-loop SL3, information flows serially from the first functional element S3F1 to the third functional element S3F3. And the third functional element S3F3 of the third sub-loop SL3 located at the end (bottom end) is connected to the third functional element MF3 of the main loop ML. In other words, the third sub-loop SL3 is connected to the main loop ML at its third functional element S3F3.

[0092] Similarly, according to MLFM, a first functional element S3F1 of the third sub-loop SL3 is provided corresponding to the function of the rotation sensor 39 in the motor control system. Second functional elements MF2, S1F2, S2F2, S3F2 of the third sub-loop SL3 are provided corresponding to the information transmission function in the motor control system. A third functional element S3F3 of the third sub-loop SL3 is provided corresponding to the function of information output (signal output) to the main loop ML in the motor control system (including the power output function).

[0093] As described above, in this embodiment, as an example of a specific disclosure of the present invention, three sub-loops SL1, SL2, and SL3 are connected to one main loop ML.

[0094] That is, in this embodiment, in the main loop ML, the multiple functional elements F are set up to the 5th (an example of N) number at the end of the main loop ML, counting from the top end along the flow of information. In addition, in the sub-loops SL1, SL2, and SL3, the multiple sub-loops SL1, SL2, and SL3 are set up to the 3rd (an example of L) number. In the first of the multiple sub-loops SL1, SL2, and SL3 (i.e., the first sub-loop SL1), the multiple functional elements F are set up to the 3rd (an example of M) number at the end of the information flow, counting from the top end of the information flow. In the second of the sub-loops SL1, SL2, and SL3 (i.e., the second sub-loop SL2), the multiple functional elements F are similarly set up to the 3rd (an example of M) number at the end of the information flow, counting from the top end of the information flow. Similarly, in the third of the sub-loops SL1, SL2, and SL3 (i.e., the third sub-loop SL3), multiple functional elements F are set up along the information flow from the top end to the third (an example of M) at the very end. As described above, in this embodiment, three sub-loops SL1, SL2, and SL3 are provided, but the present invention is not limited to this, and two or four or more sub-loops may be provided.

[0095] [...Real state acquisition part] 5 acquires the actual state quantities of each of the multiple functional elements F in the main loop ML and the first to third sub-loops SL1, SL2, and SL3. Then, for the acquired state quantities, the actual state acquisition unit 16B expresses (acquires) the state quantity U in the main loop ML as a general formula shown in the below-described formula (1), and the state quantity S in each of the sub-loops SL1, SL2, and SL3 as a vector format (matrix format) as shown in the below-described formula (2) (that is, in the present invention, the term "state quantity" can also be interpreted as a state matrix indicating each state of the functional element F).

[0096]

number

[0097]

number

[0098] The state quantity U in the main loop ML and the state quantities S in the first to third sub-loops SL1, SL2, and SL3 indicate the actual measured values ​​of the functional elements F, and are expressed in generalized vector form (matrix form) in the standard state (standard value), low state (low value), and high state (high value) relative to a predetermined reference value. Specifically, the standard value (Normal Level value) is

[0100] T It is expressed as a column vector. The low level value is

[0010] T It is expressed as a column vector. The high level value is

[0001] T It is represented as a column vector.

[0099] That is, as described above, for the main loop ML, the actual state acquisition unit 16B acquires the state quantity U i (Including the subscript U, it indicates the state quantity of the i-th functional element F in the main loop ML) is expressed in vector form (matrix form). At the same time, the actual state acquisition unit 16B expresses (acquires) a main overall state quantity V indicating the overall state quantity of the main loop ML in matrix form according to the later-described formula (3) (an example of formula 3).

[0100]

number

[0101] Incidentally, the above-mentioned formula (3) is a general formula for the main loop ML, but when this is specifically expanded to the main loop ML of this embodiment, the main overall state quantity V of the main loop ML is expressed by the formula (3)' described below.

[0102]

number

[0103] Similarly, for the first to third sub-loops SL1, SL2, and SL3, as described above, the actual state acquisition unit 16B acquires the actual state quantity S jk (Including the subscript S, it indicates the state quantity of the k-th functional element F in the j-th sub-loop SL1, SL2, SL3) is expressed in a vector form (matrix form). At the same time, the actual state acquisition unit 16B expresses each of the sub-total state quantities P j is expressed in matrix form according to equation (4) (an example of equation 1) given below.

[0104]

number

[0105] The above formula (4) is a general formula for each of the first to third sub-loops SL1, SL2, and SL3 according to the present invention. When this formula is specifically expanded (applied) to each of the first to third sub-loops SL1, SL2, and SL3, the first sub-loop SL1 is expressed by the formula (4)'-1 described below, the second sub-loop SL2 is expressed by the formula (4)'-2, and the third sub-loop SL3 is expressed by the formula (4)'-3 described below.

[0106]

number

[0107]

number

[0108]

number

[0109] [...Virtual state calculation part] The virtual state calculation unit 16C shown in FIG. 5 calculates, in a matrix format, a plurality of virtual state quantities (also understood as virtual state matrices) in the same quantity as the number of functional elements F belonging to the main loop ML or the first to third sub-loops SL1, SL2, SL3, respectively.

[0110] Regarding the main loop ML, the virtual state calculation unit 16C calculates a plurality of virtual state quantities W i is calculated according to the formula (5) shown below (an example of formula 4).

[0111]

number

[0112] Here, T shown in the above formula (5) is a virtual abnormality transfer function set between adjacent functional elements F in the main loop ML. The virtual abnormality transfer function T (also called "abnormality forward propagation matrix") is defined as a transfer function between adjacent functional elements F (nodes) as described above, and is expressed by the below-described formula (6) in this embodiment.

[0113]

number

[0114] That is, in this embodiment, the virtual abnormality transfer function T in the main loop ML and the virtual abnormality transfer functions T in the first to third sub-loops SL1, SL2, and SL3 described later are defined (set) by a unit matrix and a diagonal matrix as in the above formula (6). That is, these transfer functions are set by a unit matrix when there is no change in information or signals on the upstream side and they are transmitted to the downstream side as is. However, there are cases where information or signals are converted on the upstream side and transmitted to the downstream side. In such cases, the virtual abnormality transfer function T is appropriately and adaptively set according to (based on) the characteristics and properties of the functional element F (node).

[0115] Incidentally, the above formula (6) is a general formula for the main loop ML according to the present invention, but when this is specifically expanded (applied) to the main loop ML, the main loop ML is expressed by the formulas (5)'-1 to (5)'-5 described below.

[0116]

number

[0117]

number

[0118]

number

[0119]

number

[0120]

number

[0121] Furthermore, the virtual state calculation unit 16C calculates the above-mentioned multiple virtual state quantities W i For each of the matrix elements arranged in the horizontal direction in the formula (7) (one example of formula 4) described later, the actual state quantity S of the functional element F arranged at the end of the y-th sub-loop SL1, SL2, and SL3 is determined to be a matrix element that is a matrix element subsequent to the x-th matrix element and is calculated by matrix multiplying the virtual anomalous transfer function T. y Then, we multiply each of these by a matrix to obtain multiple virtual state quantities W i '(i=1, . . . , N (N=5 in this embodiment)) is calculated.

[0122] In this embodiment, as described above, the first sub-loop SL1 is connected to the third functional element MF3 of the main loop ML at its third functional element S1F3. The second sub-loop SL2 is connected to the first functional element MF1 of the main loop ML at its third functional element S2F3. The third sub-loop SL3 is connected to the third functional element MF3 of the main loop ML at its third functional element S3F3. Therefore, in this embodiment, the above-mentioned multiple virtual state quantities W i Each of the above is specifically expressed (expanded) as shown in equations (6)-1 to (6)-5 below.

[0123]

number

[0124]

number

[0125]

number

[0126]

number

[0127]

number

[0128] Furthermore, for each of the first to third sub-loops SL1, SL2, and SL3, the virtual state calculation unit 16C calculates a plurality of virtual state quantities Q jk is calculated according to (an example of formula 2) shown below.

[0129]

number

[0130] Here, T shown in the above formula (7) is a virtual abnormal transfer function set between the functional elements F arranged adjacently in the j-th sub-loops SL1, SL2, and SL3.

[0131] In this embodiment, the virtual abnormality transfer function T is set to be equal to T1=T2=T3 among the first to third sub-loops SL1, SL2, and SL3. The virtual abnormality transfer function T is also set to be equal to T1=T2=T3=T among the main loop ML and the sub-loops SL1, SL2, and SL3, and is set to be equal to the virtual abnormality transfer function T of the main loop ML. However, the virtual abnormality transfer function T is not limited to this, and can be appropriately and adaptively set according to specific cases such as the specific characteristics and properties of the functional element F and / or the case where information or a signal is converted on the upstream side and transmitted to the downstream side, as described above.

[0132] Here, the above-mentioned formula (7) is a general formula for the sub-loops SL1, SL2, and SL3 according to the present invention. When this formula is specifically expanded (applied) to the first to third sub-loops SL1, SL2, and SL3 of this embodiment, the multiple virtual state quantities Q 11 ~Q 13 ,Q 21 ~Q 23 ,Q 31 ~Q 33 are expressed (expanded) by the following equations (7)'-1-1 to (7)'-1-3, (7)'-2-1 to (7)'-2-3, and (7)'-3-1 to (7)'-3-3, respectively.

[0133]

number

[0134]

number

[0135]

number

[0136]

number

[0137]

number

[0138]

number

[0139]

number

[0140]

number

[0141]

number

[0142] [...Status comparison section] The state comparison unit 16D shown in FIG. 5 compares the above-mentioned multiple virtual state quantities W i ',Q jk Each of these is compared with the results acquired by the actual state acquisition unit 16B.

[0143] Specifically, for the main loop ML, the state comparison unit 16D compares a plurality of virtual state quantities W i ' (see above) and judge whether they match. At that time, in the anomaly detection system 1, the virtual state quantity W i ' is set as corresponding to the i-th (i=1, . . . , N (N=5 in this embodiment)) functional element F in the main loop ML.

[0144] For the first to third sub-loops SL1, SL2, and SL3, the state comparison unit 16D of the server device 10 calculates a plurality of virtual state quantities Q jk (see above) for each of the sub-total state quantities P j In the anomaly detection system 1, the virtual state quantity Q jk is set to correspond to the k-th (k=1,...,M (M=3 in this embodiment)) functional element F in the j-th (j=1,...,L (L=3 in this embodiment)) sub-loop SL1, SL2, SL3.

[0145] [...Abnormality identification part] The abnormality identifying section 16E of the server device 10 shown in FIG. 5 identifies the functional element F in which an abnormality has occurred based on the comparison result of the above-mentioned state comparing section 16D.

[0146] Specifically, for the main loop ML, when only one match is found as a result of the comparison by the state comparison unit 16D, the abnormality identification unit 16E of the server device 10 identifies the matched virtual state quantity W i Based on the virtual state quantity W i Alternatively, if there are multiple matches, the abnormality identification unit 16E of the server device 10 identifies the i-th functional element F corresponding to the virtual state quantity W i Based on this, the virtual state quantity W i Among the corresponding functional elements F of ', the functional element F arranged at the top is identified. If there is no match in the main loop ML, the abnormality identifying unit 16E of the server device 10 determines that no abnormality has occurred in the main loop ML.

[0147] Then, for each of the first to third sub-loops SL1, SL2, and SL3, if there is only one match in the j-th sub-loop SL1, SL2, or SL3 as a result of the comparison by the state comparison unit 16D described above, the abnormality identification unit 16E of the server device 10 identifies the matched virtual state quantity Q jk Based on the virtual state quantity Q jk Alternatively, if there are multiple matches in the j-th sub-loops SL1, SL2, and SL3, the abnormality identification unit 16E of the server device 10 identifies the k-th functional element F corresponding to the j-th sub-loops SL1, SL2, and SL3. jk Based on this, the virtual state quantity Q jk Among the corresponding functional elements F, the functional element F arranged at the top is identified.

[0148] In addition, if there is no match in the first to third sub-loops SL1, SL2, and SL3, the abnormality identification unit 16E of the server device 10 determines that no abnormality has occurred in the first to third sub-loops SL1, SL2, and SL3.

[0149] [...Abnormality detection section] 5 detects an abnormality in the function of the motor control system based on the identification result by the abnormality identification unit 16E described above. That is, in this embodiment, the abnormality detection unit 16F of the server device 10 detects that an abnormality has occurred in the function of the motor control system (hardware such as an electric circuit, component, or device in the real world) corresponding to the functional element F identified by the abnormality identification unit 16E of the server device 10. Then, the abnormality detection unit 16F of the server device 10 transmits the detection result to the information terminal device 20 via the communication unit 18 of the server device 10.

[0150] If it is determined that no abnormality has occurred in any of the main loop ML and the first to third sub-loops SL1, SL2, and SL3, the abnormality identification unit 16E of the server device 10 transmits information to that effect (information of a normal state) to the main body of the control unit 16 of the server device 10. The control unit 16 of the server device 10 loops and executes the series of programs (flows) described above (see below) as long as no abnormality is detected, that is, as long as there is no match in any of them.

[0151] [Information terminal device] As shown in FIG. 5, the information terminal device 20 of the user H includes a communication section 28, a storage section 27, an operation input section 23, an image output section 24, and a control section 26. The communication unit 28 of the information terminal device 20 also communicates with the server device 10 in the same manner as the server device 10 .

[0152] The storage unit 27 of the information terminal device 20 stores and retains information required by the information terminal device 20, and the storage unit 27 of the information terminal device 20 at least temporarily stores and retains control signal information exchanged within the vehicle steering device 30 as described above. The operation input unit 23 of the information terminal device 20 accepts operations from a user H who operates the information terminal device 20. The image output unit 24 of the information terminal device 20 displays at least information on the abnormality detection in the information terminal device 20 according to control from the server device 10.

[0153] The control unit 26 of the information terminal device 20 has a request sending unit 26A, a response receiving unit 26B, and an abnormality detection control unit 26C, and executes processing relating to abnormality detection in the information terminal device 20.

[0154] The request sending unit 26A of the information terminal device 20 sends a request to the server device 10 through the communication unit 28 of the information terminal device 20 based on operation information received by the operation input unit 23 of the information terminal device 20 from the user H. The response receiving unit 26B of the information terminal device 20 receives a response such as a processing result related to anomaly detection in response to the request sent by the request sending unit 26A to the server device 10.

[0155] The abnormality detection control unit 26C of the information terminal device 20 adaptively controls the storage unit 27, operation input unit 23, image output unit 24, etc. of the information terminal device 20 through transmission by the request transmission unit 26A and reception by the response reception unit 26B to obtain and notify information for abnormality detection. The abnormality detection control unit 26C of the information terminal device 20 also controls and executes the transmission and reception of the information to and from the server device 10.

[0156] 5 may be configured to realize some of the functions of the server device 10 and the information terminal device 20. Also, the collection of various functions of the server device 10 and the information terminal device 20 shown in FIG. 5 can be interpreted as one information processing device or information processing system from a global perspective. How multiple functions necessary to realize the present invention are allocated to one or more pieces of hardware and / or software is appropriately determined based on the processing capacity of each piece of hardware, the structure and configuration of the program, or the specifications required for the anomaly detection system 1.

[0157] [About the operation flow executed on the server device] An operational flow regarding anomaly detection executed by the server device 10 will be described with reference to FIG. FIG. 7 is a flow diagram illustrating an example of an operation flow regarding anomaly detection executed by the server device 10 shown in FIG.

[0158] As shown in FIG. 7, the model setting unit 16A of the server device 10 provides multiple functional elements F that individually correspond to each of the functions of the motor control system according to the MLFM (multi-level flow model) method, and sets one main loop ML and multiple (in this embodiment, first to third) sub-loops SL1, SL2, and SL3 (see FIG. 6, S1).

[0159] The real state acquisition unit 16B of the server device 10 acquires the real state quantities of each of the multiple functional elements F in the main loop ML and the first to third sub-loops SL1, SL2, and SL3 (S2). At this time, as described above, the real state acquisition unit 16B of the server device 10 expresses the state quantity U in the main loop ML and the state quantities S in the first to third sub-loops SL1, SL2, and SL3 in a vector format (matrix format). At the same time, the real state acquisition unit 16B of the server device 10 expresses each of these overall state quantities as the main overall state quantity V and the sub overall state quantity P. j It is expressed (defined) in the form of a matrix (j=1, . . . , L (L=3 in this embodiment)).

[0160] The virtual state calculation unit 16C of the server device 10 calculates, in a matrix format, a number of virtual state quantities equal to the number of functional elements F belonging to the main loop ML or the first to third sub-loops SL1, SL2, SL3, respectively (S3).

[0161] Regarding the main loop ML, the virtual state calculation unit 16C of the server device 10 calculates a plurality of virtual state quantities W iFor each of the first to third sub-loops SL1, SL2, and SL3, the virtual state calculation unit 16C of the server device 10 calculates a plurality of virtual state quantities Q jk (j=1, . . . , L (L=3 in this embodiment); k=1, . . . , M (M=3 in this embodiment)) is calculated.

[0162] The state comparison unit 16D of the server device 10 compares the multiple virtual state quantities W calculated by the virtual state calculation unit 16C. i ',Q jk The state comparison unit 16D of the server device 10 first executes a process (state comparison) related to the main loop ML.

[0163] Regarding the main loop ML, the state comparison unit 16D of the server device 10 compares the multiple virtual state quantities W i The main overall state quantity V is compared with each of the main loop ML' (see above) to determine whether they match (S5). If it is determined that there is a match in the main loop ML as a result of the match determination (YES in S5), the abnormality identification unit 16E of the server device 10 determines whether the number of matches in the main loop ML is one (S6). If it is determined that there is no match in the main loop ML (NO in S5), the operation flow proceeds to step S9.

[0164] As a result of judging whether the number of matches in the main loop ML is one or not (S6), if it is judged that the number of matches is one (YES in S6), the abnormality identification unit 16E of the server device 10 i Based on the virtual state quantity W i On the other hand, if there is only one mismatch (NO in S6), that is, if there are multiple matches, the anomaly identification unit 16E of the server device 10 identifies the i-th functional element F corresponding to the virtual state quantities W iBased on this, the virtual state quantity W i Among the functional elements F corresponding to each of the groups, the functional element F arranged at the top is identified (S8).

[0165] After completing the process related to the main loop ML, the state comparison unit 16D of the server device 10 then executes the processes related to the first to third sub-loops SL1, SL2, and SL3 (S9).

[0166] Regarding the first to third sub-loops SL1, SL2, and SL3, the state comparison unit 16D of the server device 10 calculates a plurality of virtual state quantities Q jk For each of the sub-global state quantities P j and determine whether or not there is a match (S10). If it is determined that there is a match in the first to third sub-loops SL1, SL2, and SL3 as a result of the match determination (YES in S10), the abnormality identification unit 16E of the server device 10 determines whether or not the number of matches in the first to third sub-loops SL1, SL2, and SL3 is one (S11). If it is determined that there is no match in the first to third sub-loops SL1, SL2, and SL3 (NO in S11), the operation flow proceeds to step S14.

[0167] As a result of judging whether the number of matches in the first to third sub-loops SL1, SL2, and SL3 is one or not (S11), if it is judged that the number of matches is one (YES in S11), the abnormality identification unit 16E of the server device 10 detects the matched virtual state quantity Q jk Based on the virtual state quantity Q jk On the other hand, if there is only one mismatch (NO in S11), that is, if there are multiple matches, the abnormality identification unit 16E of the server device 10 identifies the k-th functional element F corresponding to the multiple virtual state quantities Q jk Based on this, the virtual state quantity Q jk Among the corresponding functional elements F, the functional element F arranged at the top is identified (S13).

[0168] The abnormality detection unit 16F of the server device 10 detects whether or not an abnormality has occurred in the function of the motor control system corresponding to the functional element F identified by the abnormality identification unit 16E of the server device 10 (S14). At the same time, the abnormality detection unit 16F of the server device 10 determines whether or not an abnormality has occurred in any of the main loop ML and the first to third sub-loops SL1, SL2, and SL3 (S15).

[0169] As a result of the determination of whether or not an abnormality has been detected (S15), if no abnormality is detected (NO in S15), the operation flow returns to step S2. On the other hand, if it is determined that an abnormality has been detected (YES in S15), the abnormality detection unit 16F of the server device 10 transmits the detection result to the information terminal device 20 through the communication unit 18 of the server device 10 (S16). More specifically, the server device 10 causes the information terminal device 20 to display the occurrence of the abnormality and its location as abnormality information on its image output unit 24 through the communication unit 18. This enables the server device 10 to notify the user H of the vehicle C of the abnormality information through the information terminal device 20.

[0170] [Advantages and features of this embodiment] As described above, the anomaly detection system 1 of this embodiment is connected in hardware to a specific motor control system (an example of a control system) that is actually operating and is configured to include an electric circuit system (an example of a circuit system), and the control system is modeled in software to detect an anomaly in the control system. A model setting unit 16A sets one main loop ML (an example of a main group) represented by a plurality of functional elements F connected in series with each other and through which information flows serially from their upper ends to their end ends, and one or more sub-loops SL1, SL2, SL3 (an example of a sub-group) represented by a plurality of functional elements F connected directly with each other and through which information flows serially from their upper ends to their end ends, and each of the sub-loops SL1, SL2, SL3 is connected to any one of the plurality of functional elements F of the main loop ML at its endmost functional element F; an actual state acquisition unit 16B acquires actual state quantities V, P, U, S of the plurality of functional elements F in the main loop ML and the plurality of sub-loops SL1, SL2, SL3, respectively, and expresses the state quantities V, P, U, S in a matrix format; and a plurality of virtual state quantities W i ',Q jk A virtual state calculation unit 16C calculates a plurality of virtual state quantities W i ',Q jk For each of the above, a state comparison unit 16D is included which compares the result of the state comparison unit 16B with the result of the state comparison unit 16B, and an abnormality identification unit 16E is included which identifies the functional element F in which an abnormality has occurred based on the comparison result of the state comparison unit 16D.

[0171] Furthermore, according to the anomaly detection method of this embodiment, a specified device and / or program is used to connect in hardware to a specified motor control system (an example of a control system) that is actually operating and includes an electrical circuit system (an example of a circuit system), and the control system is modeled in software to detect an abnormality in the control system. a model setting step of setting one main loop ML (an example of a main group) represented by a plurality of functional elements F connected in series with each other and through which information flows serially from their upper ends to their end ends, and one or more sub-loops SL1, SL2, SL3 (an example of a sub-group) represented by a plurality of functional elements F connected directly with each other and through which information flows serially from their upper ends to their end ends, each of which is connected to one of the plurality of functional elements F of the main loop ML at its endmost functional element F; an actual state acquisition step of acquiring actual state quantities V, P, U, S of the plurality of functional elements F in each of the main loop ML and the plurality of sub-loops SL1, SL2, SL3 and expressing the state quantities V, P, U, S in a matrix format; and a plurality of virtual state quantities W in each of the main loop ML and the plurality of sub-loops SL1, SL2, SL3, the number of which is the same as the number of functional elements F belonging to the main loop ML or the sub-loops SL1, SL2, SL3. i ',Q jk A virtual state calculation step of calculating a virtual state quantity W i ',Q jk For each of the above, it includes a state comparison step of comparing the result acquired in the actual state acquisition step with the actual state, and an abnormality identification step of identifying the functional element F in which an abnormality has occurred based on the comparison result in the state comparison step.

[0172] For this reason, a multi-level flow model modeling technique is adopted, and even if the control system is complex and the changes over time are large compared to, for example, a plant, it is possible to quickly detect abnormalities in the control system and identify and visualize the location of the abnormality.

[0173] The invention will now be described in further detail with reference to one or more examples, which are not intended to limit the subject matter recited in the claims. EXAMPLES

[0174] In the present embodiment, the present invention will be described in more detail by taking a number of examples as specific examples of the present invention.

[0175] <First Example> A first embodiment of the present invention will be described. In this embodiment, an example will be described assuming that an abnormality occurs in the second sub-loop SL2 in the first embodiment. Specifically, a case is assumed in which the first functional element S2F1 of the second sub-loop SL2 fails and outputs a high level value.

[0176] The actual state acquisition unit 16B of the server device 10 acquires (expresses) a sub-total state quantity P2 indicating the overall actual state quantity of the second sub-loop SL2 as shown in equation (14)'-2 below, for the second sub-loop SL2.

[0177]

number

[0178] At that time, the virtual state calculation unit 16C of the server device 10 sets the virtual abnormality transfer function T as a unit matrix, and calculates each of the virtual state quantities related to the second sub-loop SL2 as shown in the later-described equations (17)'-2-1 to (17)'-2-3.

[0179]

number

[0180]

number

[0181]

number

[0182] The state comparison unit 16D of the server device 10 compares the plurality of virtual state quantities Q 21 ,Q 22 ,Q 23 The sub-total state quantity P2 is compared with each of them to determine whether they match.

[0183] At that time, as described above, in the anomaly detection system 1, the virtual state quantity Q 21 is the first functional element S2F1 of the second sub-loop SL2, and the virtual state quantity Q 22 is the second functional element S2F2 of the second subloop SL2, and the virtual state quantity Q 23 is set as corresponding to the third functional element S2F3 of the second sub-loop SL2.

[0184] In this embodiment, the virtual state quantity Q 21 ,Q 22 ,Q 23 Each of these is the same as the sub-total state quantity P2, and the state comparison unit 16D of the server device 10 judges that both of them match.

[0185] The abnormality identification unit 16E of the server device 10 identifies the functional element F in which the abnormality has occurred based on the comparison result of the state comparison unit 16D. In the present embodiment, all three match, so the matched virtual state quantities Q 21 ,Q 22 ,Q 23 Based on this, the virtual state quantity Q 21 ,Q 22 ,Q 23 Among the functional elements F corresponding to each of the sub-loops SL2 and SL3, the functional element F arranged at the top is identified. That is, the abnormality identifying unit 16E of the server device 10 identifies the first functional element S2F1 of the second sub-loop SL2 (see FIG. 6).

[0186] Based on the identification result by the abnormality identification unit 16E, the abnormality detection unit 16F of the server device 10 detects a failure at a location corresponding to the first functional element S2F1 of the second sub-loop SL2 (i.e., the battery 41) as an abnormal location in the function of the motor control system. The abnormality detection system 1 notifies the user H of the abnormal state (abnormality information) by displaying the detection result on the image output unit 24 of the information terminal device 20 through the communication unit 18 of the server device 10.

[0187] <Second Example> A second embodiment of the present invention will be described. In this embodiment, an example will be described assuming that the first to third sub-loops SL1, SL2, and SL3 are normal in the first embodiment, but an abnormality occurs in the main loop ML. Specifically, a case is assumed in which the third functional element MF3 of the main loop ML fails and outputs a Low Level value.

[0188] The actual state acquisition unit 16B of the server device 10 acquires, with respect to the main loop ML, a main overall state quantity V that indicates the overall actual state quantity of the main loop ML as shown in equation (13)' below.

[0189]

number

[0190] That is, in this embodiment, the first and second functional elements MF1, MF2 of the main loop ML are in a state of outputting a high level value, and the third to fifth functional elements MF3, MF4, MF5 are in a state of outputting a low level value.

[0191] In this embodiment, the first to third sub-loops SL1, SL2, and SL3 are assumed to be in a normal state, so the actual state quantity S 13 , the actual state quantity S of the third functional element S2F3 arranged at the end of the second sub-loop SL2 23, and the actual state quantity S of the third functional element S3F3 arranged at the end of the third sub-loop SL3. 33 Each of the above is expressed (shown) by the following equations (22)-1 to (22)-3.

[0192]

number

[0193]

number

[0194]

number

[0195] Therefore, the virtual state calculation unit 16C of the server device 10 sets the virtual abnormality transfer function T as a unit matrix, and calculates each of the virtual state quantities W1', W2', W3', W4', and W5' related to the main loop ML as shown in Equations (26)-1 to (26)-5 described below.

[0196]

number

[0197]

number

[0198]

number

[0199]

number

[0200]

number

[0201] The state comparison unit 16D of the server device 10 compares the main entire state quantity V with each of the plurality of virtual state quantities W1', W2', W3', W4', and W5' related to the main loop ML to determine whether they match.

[0202] At this time, as described above, in the anomaly detection system 1, the virtual state quantity W1' is set to correspond to the first functional element MF1 of the main loop ML, the virtual state quantity W2' is set to correspond to the second functional element MF2 of the main loop ML, the virtual state quantity W3' is set to correspond to the third functional element MF3 of the main loop ML, the virtual state quantity W4' is set to correspond to the fourth functional element MF4 of the main loop ML, and the virtual state quantity W5' is set to correspond to the fifth functional element MF5 of the main loop ML.

[0203] In this embodiment, each of the virtual state quantities W3', W4', and W5' is identical to the main overall state quantity V, and the state comparison unit 16D of the server device 10 determines that the three virtual state quantities W3', W4', and W5' match.

[0204] The abnormality identifying unit 16E of the server device 10 identifies the functional element F in which an abnormality has occurred based on the comparison result of the state comparing unit 16D. In the present embodiment, since there are three matches, the uppermost functional element F among the functional elements F corresponding to each of the virtual state quantities W3', W4', and W5' is identified based on the matching virtual state quantities W3', W4', and W5'. That is, the abnormality identifying unit 16E of the server device 10 identifies the third functional element MF3 of the main loop ML (see FIG. 6).

[0205] Based on the identification result by the abnormality identification unit 16E, the abnormality detection unit 16F of the server device 10 detects a failure at a location corresponding to the third functional element MF of the main loop ML (i.e., the ECU 40) as an abnormal location in the function of the motor control system. The abnormality detection system 1 notifies the user H of the abnormal state (abnormal information) by displaying the detection result on the image output unit 24 of the information terminal device 20 via the communication units 18, 28 of the server device 10.

[0206] <Conclusion> Although the specific embodiments have been described above, the aspects of the present invention are not limited to these embodiments, and suitable modifications and improvements are possible. [Industrial Applicability]

[0207] The present invention is useful as an anomaly detection device and an anomaly detection method that employ a multilevel flow model modeling technique and can quickly detect anomalies in a control system and identify and visualize the location of the anomaly, even when the control system is complex and the changes over time are large compared to, for example, a plant. [Explanation of symbols]

[0208] 1: Anomaly detection system 10: Server equipment 11: CPU 12:Storage device 13: Operation input device 14: Image output device 15: Communication equipment 16: Control section 16A: Model setting section 16B: Reality state acquisition unit 16C: Virtual state calculation unit 16D: Status comparison section 16E: Abnormality identification part 16F: Anomaly detection section 17: Storage section 18: Communications Department 20: Information terminal device 23: Operation input section 24: Image output section 26: Control section 26A: Request sending section 26B: Response receiving section 26C: Abnormality detection control unit 27: Storage section 28: Communications Department 30: Vehicle steering system 31: Steering shaft 32: Reduction gear 33: Universal joint 34: Pinion rack mechanism 34A: Pinion 34B: Rack 35: Tie rod 36: Hub unit 37: Steering wheel 38: Motor 39: Rotation sensor 40: ECU 41: Battery BS: Bus C: Vehicle H: User ML: Main Loop MF1: ​​First functional element MF2: Second functional element MF3: The third functional element MF4: The fourth functional element MF5: The fifth functional element N: Communication network S1F1: First functional element S1F2: Second functional element S1F3: The third functional element S2F1: First functional element S2F2: Second functional element S2F3: The third functional element S3F1: First functional element S3F2: Second functional element S3F3: The third functional element SL1: First sub-loop SL2: Second sub-loop SL3: 3rd sub-loop

Claims

1. An anomaly detection system that is connected in a hardware manner to a predetermined control system that is configured including a circuit system and actually operates, models the control system in software, and detects anomalies in the control system, a model setting unit which provides a plurality of functional elements corresponding to the respective functions of the control system in accordance with a multilevel flow model technique, and sets one main group represented by the plurality of functional elements connected in series with each other and through which information flows serially from its upper end to its end, and one or more subgroups represented by the plurality of functional elements connected directly with each other and through which information flows serially from its upper end to its end, each of which is connected to any one of the plurality of functional elements of the main group at its endmost functional element; an actual state acquisition unit that acquires actual state quantities of the functional elements in the main group and each of the subgroups and expresses the actual state quantities in a matrix format; a virtual state calculation unit that calculates, in a matrix format, a number of virtual state quantities in the main group and each of the plurality of subgroups, the same number as the number of the functional elements belonging to the main group or each of the plurality of subgroups; a state comparison unit that compares each of the plurality of virtual state quantities calculated by the virtual state calculation unit with an acquired result by the real state acquisition unit; and an abnormality identification unit that identifies the functional element in which an abnormality has occurred based on a comparison result of the state comparison unit. Anomaly detection system.

2. In the case where the plurality of subgroups are set up to an L-th subgroup, and the plurality of functional elements are set in a j-th subgroup (j is a natural number; j=1, . . . , L) from the top end along the information flow to an M-th subgroup (M is a natural number), The actual state acquisition unit acquires the actual state quantity S of the k-th (k is a natural number; k=1, . . . , M) functional element in the j-th subgroup. jk (indicating the state quantity of the k-th functional element in the j-th subgroup) in vector form, and a sub-whole state quantity P j is expressed in matrix form according to Equation 1 below, The virtual state calculation unit calculates the multiple virtual states Q jk is calculated according to the following formula 2, The state comparison unit compares the plurality of virtual states Q jk For each of the sub-groups, the sub-total state quantity P j to determine whether they match, The anomaly detection system according to claim 1 . [Formula 1] P j =(S j1 S j2 ・・・ S j(M-1) S jM ) where j is a natural number; j=1, . . . , L [Formula 2] Q jk =(S j1 ・・・ S j(k-1) S jk T 1 1 ・S jk ・・・ T 1 M-k ・S jk ) where j and k are both natural numbers; j=1,...,L; k=1,...,M T j is a virtual abnormal transfer function set between the functional elements that are adjacently arranged in the j-th subgroup.

3. an abnormality detection unit that detects an abnormality in a function of the control system based on a result of the identification by the abnormality identification unit, The virtual state quantity Q jk is set to correspond to the k-th functional element in the j-th subgroup, When there is only one match in the j-th subgroup as a result of the comparison by the state comparison unit, the anomaly identification unit identifies the matched virtual state quantity Q jk Based on the virtual state quantity Q jk or, if there are a plurality of coincident virtual state quantities Q jk Based on this, the virtual state quantity Q jk identifying the functional element arranged at the top among the functional elements corresponding to the The abnormality detection unit detects that an abnormality has occurred in a function of the control system corresponding to the functional element identified by the abnormality identification unit. The anomaly detection system according to claim 2 .

4. In the main group, the plurality of functional elements are counted from the top to the end of the Nth functional element (N is a natural number) along the information flow, The actual state acquisition unit acquires the state quantity U of the i-th (i is a natural number; i=1, . . . , N) functional element in the main group. i (indicating the state quantity of the i-th functional element in the main group) in vector form, and expressing a main overall state quantity V indicating the state quantity of the entire main group in matrix form according to Equation 3 described later; The virtual state calculation unit calculates the plurality of virtual state quantities W i is calculated according to the following formula 4, Furthermore, the virtual state calculation unit calculates the virtual state quantities W i For each of the matrix elements arranged in the horizontal direction in Equation 4 described later, the actual state quantity S of the functional element arranged at the end of the y-th subgroup is determined to be a matrix element that is the next matrix element after the x-th matrix element and is calculated by matrix multiplying a virtual transfer function T described later. y Further, by matrix multiplying each of these, multiple virtual state quantities W i ' (i=1, . . . , N), The state comparison unit compares the plurality of virtual state quantities W i ' to determine whether or not the main overall state quantity V coincides with each other. The anomaly detection system according to claim 1 . [Formula 3] V = (U 1 U 2 ・・・ U N-1 U N ) [Formula 4] W i = (U 1 ・・・ U i-1 U i T 1 ・U i ・・・ T N-i ・U i ) where i is a natural number; i=1, . . . , N T is a virtual abnormal transfer function set between the functional elements that are adjacently arranged in the main group.

5. an abnormality detection unit that detects an abnormality in a function of the control system based on a result of the identification by the abnormality identification unit, The virtual state quantity W i ' is set to correspond to the i-th functional element in the main group, When only one virtual state quantity W matches as a result of the comparison by the state comparison unit, the abnormality specification unit i Based on the virtual state quantity W i ', or, if there are a plurality of coincident virtual state quantities W i Based on this, the virtual state quantity W i ', identify the functional element that is arranged at the top among the functional elements corresponding to The abnormality detection unit detects that an abnormality has occurred in a function of the control system corresponding to the functional element identified by the abnormality identification unit. The anomaly detection system according to claim 4 .

6. The virtual anomalous transfer function is an identity matrix. The anomaly detection system according to claim 2 or 4.

7. An anomaly detection method for detecting an anomaly in a control system that is actually operating and includes a circuit system, by connecting the control system in a hardware manner using a predetermined device and / or program, and modeling the control system in software, comprising: a model setting step of providing a plurality of functional elements corresponding to each of the functions of the control system in accordance with a multilevel flow model technique, and setting one main group represented by the plurality of functional elements connected in series with each other and through which information flows serially from the upper end to the lower end, and one or more subgroups represented by the plurality of functional elements connected directly with each other and through which information flows serially from the upper end to the lower end, each of which is connected to any one of the plurality of functional elements of the main group at its most distal functional element; an actual state acquisition step of acquiring actual state quantities of the respective functional elements in the main group and the respective subgroups and expressing the actual state quantities in a matrix format; a virtual state calculation step of calculating, in a matrix format, a number of virtual state quantities equal to the number of the functional elements belonging to the main group or the subgroups, in each of the main group and the subgroups; a state comparison step of comparing each of the plurality of virtual state quantities calculated in the virtual state calculation step with an acquired result in the actual state acquisition step; and an abnormality identification step of identifying the functional element in which an abnormality has occurred based on a comparison result in the state comparison step. Anomaly detection methods.

Images