To explore the relationship between job satisfaction and occupational stress in the workers of a thermal power plant. The cluster sampling method was used to investigate 875 workers in a thermal power plant. The job satisfaction, occupational stressors, strains, personalities, meeting strategy and social support were measured using occupational stress instruments, job content questionnaire and effort-reward imbalance questionnaire. There were no significant differences of job satisfaction scores between different groups according to sex, educational level, marriage status, smoking and drinking (P > 0.05). But there were significant differences of job satisfaction scores between different age groups or between different service length groups (P < 0.01). The correlation analysis revealed that job satisfaction scores were related positively to responsibility for persons and things, promotion opportunity, job control, job stabilization, rewards, mental health, positive affectivity, self-esteem, superior support and coworker support scores (P < 0.05 or P < 0.01), negatively to role ambiguity and conflict, job future ambiguity, job demands, negative affectivity, depressive symptoms, patience, and mental locus of work control(P < 0.05 or P < 0.01). The results of variance analysis indicated that the job relationship, responsibility for persons and things, promotion opportunity, job control, job stabilization, reward, mental health, positive affectivity, self-esteem and buffer scores of the workers with high job satisfaction scores were significantly higher than those of workers with moderate and lower job satisfaction scores (P < 0.01), but the role ambiguity and conflict, job future ambiguity, job demands, and depressive symptoms scores and mental locus of work control of the workers with high job satisfaction scores were significantly lower than those of workers with moderate and lower job satisfaction scores (P < 0.05 or P < 0.01). The results of multivariate logistic regression analysis showed that the risk of job dissatisfaction for workers with low reward was about four times as high as that for workers with high reward (OR = 3.773), the risks of job dissatisfaction for workers with low social support and mental locus of external work control were about two times as high as that for workers with high social support or mental locus of internal work control (OR = 2.419 and 2.219, respectively). The daily life stress, low control strategy, low support strategy, low job control and negative affectivity were risk factors of job dissatisfaction (OR = 1.125 approximately 1.790), but the self-esteem and positive affectivity could reduce the risk of job dissatisfaction. Increasing the decision level, social support, meeting strategy and reward or decreasing the role conflict and ambiguity, demands and negative affectivity could improve worker's job satisfaction level.