Hitachi, Ltd.

Hitachi Engineering Co., Ltd.

Haruki Inoue

Masakazu Yahiro

Yoshiyuki Satoh

Motohisa Funabashi

Yoshiki Kobayashi

Kenichi Nakamura

Masao Takatoo

Tadaaki Kitamura

Kazuyoshi Asada

Kunio Nakanishi

Toshiro Shibata

Takeshi Horie

Hideo Yoshida

Noboru Abe

Katsuyuki Yamamoto

Mayumi Mizutani

Masami Shiozawa

Keiji Oshima

Hiroyuki Ichikawa

Hitoshi Onizawa

Minoru Koide

Yukio Hamaguchi

Yasuhiro Terada

Fumiki Tanaka

Akemi Ohtsuki

Satoru Yoshikawa

A planning system includes an input device which accepts at least a given planning problem and values of variables required to solve the given planning problem; a display device which provides a visual display of an optimum plan obtained for the given planning problem; and a planner device which prepares an objective function representing items to be averaged and designs a plan to minimize a value of the objective function. The planner device includes an initial plan reader which reads an initial plan; an objective function value operation unit which calculates an objective function value of the initial plan; a reproduction unit which generates random numbers to a set composed by elements contacting to boundary planes of divided subsets, wherein the random numbers include a range of number of contacting elements and an equally-opportunity-selectable distribution, select at least one element based on the random numbers, equal-opportunity-selecting one of the subsets to which element contacts, and schedule the plan such that the selected element is rendered to belong to the selected subset; a plan renewal unit which compares the objective function value of the current plan with the objective function value of the previous plan to determine the plan having a smaller objective function value as an optimum plan; and a control unit which provides a visual display of an optimum plan on the display device.

Planning system and planning method

An object recognition system using the image processing in which an area having a unique feature is extracted from an input image of an object, the unique image is registered in a shade template memory circuit as a shade template, the input image is searched for an image similar to the shade template registered by a shade pattern matching circuit, the position of an object is determined for each template, the speed and direction of movement of the object is determined from the positional information, and the results thereof are integrated by a separation/integration circuit, thereby recognizing the whole of the moving object.

Object recognition system and abnormality detection system using image

A system for making an optimum plan for a given problem at high speed. The system sets a planning problem, etc., prepares an objective function and finalizes a plan for minimizing or maximizing the objective function value, and then stores necessary variables. The system prepares as many parent plans of a first generation as a given number (population), calculates objective function values, sorts plans in descending or ascending order of the objective function values of the plans, when selection numbers are assigned to plans in order of a ratio of the objective function value of each plan to the total of the objective function values of the plans, selects parent plans by using numbers indicated by as many constants as the population determined for each generation as the selection numbers, and replaces two elements, arranged at order positions specified by random numbers, with each other for each of the selected parent plans, to prepare child plans.

Planning method and system

A planning system for quickly forming the optimum plan for a given planning problem, includes a setting unit for accepting the given planning problem; an optimization unit for creating an objective function which expresses an item intended to be minimized or maximized in the planning problem, and for executing a process which minimizes or maximizes a value of the created objective function; a storage unit for storing therein, at least, constant values which are required for the optimizing process; and a display unit for displaying a processed result of the optimization unit. The optimization unit performs a first function of calculating a difference value between the values of the objective function assumed in a last formed plan and a newly formed plan, and performs a second function of comparing the difference value of the objective function with the constant value set in the storage unit, and then substituting the last formed plan by the newly formed plan and setting the latter as a candidate for an optimum plan, on condition that the difference value of the objective function is smaller than the constant value.

A neural network system includes a qualitative evaluation section, a neural network section, a quantifying section and a display section. The qualitative evaluation section qualitatively analyzes an unknown data supplied thereto, and normalizes the result of analysis within a predetermined range. The neural network section having a neural network with plural neurons computes the network output data from the normalized unknown data produced by the qualitative evaluation section. Each neuron is connected to plural other neurons through synapses, each of which is assigned an individual weight coefficient. Each neuron is adapted to output an output function value assigned thereto associated with the total sum of the products of the output from the neurons connected thereto and the synapse weight coefficient. The quantifying section quantifies the network output data to produce desired data. The desired data thus produced is displayed on the display section.

Neural network system adapted for non-linear processing

A process control system for controlling a process exhibiting both linear behavior and non-linear behavior with a plurality of control quantities has a knowledge base storing therein universal facts, production rules including expert's empirical rules, meta rules describing flows of inferences, algorithm methods by mathematical expressions and membership functions referenced in fuzzy inference and a complex inference mechanism. The complex inference mechanism is composed of a complex fuzzy mechanism for inferring directly predicted values on a multi-dimensional space generated by a plurality of elements for evaluation and predicted control objectives evaluated previously with fuzzy quantities, a predicting fuzzy mechanism having the input supplied with the predicted values for determining arithmetically correlation to the membership functions determined previously with the fuzzy quantities and a satisfaction grade obtained by imparting weight to the correlation for each of the control objectives to thereby determine such a combination of the control quantities for control effectors to vary the operation states thereof from the current operation states that the satisfaction grade becomes maximum, and a main inference mechanism having the input supplied with process data to thereby compare selectively the process data with the knowledge stored in the knowledge base for thereby making decision of the process behavior through a forward fuzzy inference with the aid of the production rules based on the empirical knowledge and manage the whole process control system including the complex fuzzy inference mechanism, the predicting fuzzy inference mechanism and the knowledge base. The system can be advantageously employed for intra-tunnel ventilation control and other processes exhibiting both linear and non-linear behaviors.

Method and system for process control with complex inference mechanism

A prediction control of a process containing a non-linear behavior is effected by predicting a transition of process quantities (control quantities) of an object to be controlled a predetermined time period after the current time to provide predicted values and by determining manipulation quantities of at least a control effector in accordance with a difference between the predicted values and predetermined target values of the controlled object. A quantitative operation is carried out arithmetically identifying a process as a linear behavior, and a fuzzy inference or qualitative operation, including a fuzzy rule based on empirical knowledge, is carried out for simulating a process, wherein for input process information and current manipulation quantities to be maintained, both operations parallelly determine predicted values of control quantities. A process behavior determination is arithmetically carried out to determine whether the process linearly behaves (i.e. the normal state or the state dominant to statistic distribution) or not (i.e. the extreme transition state such as a transient response, or an indeterminate state) from the specified process information on the basis of a process behavior determination rule. As a result, a predicted value acquired by the quantitative operation is selected in response to a determination of a linear behavior, and a predicted value acquired by the fuzzy inference operation is selected in response to a determination of a non-linear behavior.

A process control system for controlling actions of a process is provided as a qualitative model to which the fuzzy theory is applied to infer and control the process. The process control system includes differentiation apparatus for determining the amounts of displacement of different process amounts with respect to time, qualitative fuzzy inference apparatus for making inference of qualitative process actions, and centroid arithmetic apparatus for converting the qualitative values of inferred control target process amounts into the appropriate amount of control.

Process control method and control system

Growing community of inventors

Katsuta, Japan

Haruki Inoue